Census of the Local Universe (CLU) Narrow-Band Survey I: Galaxy Catalogs from Preliminary Fields

Mon. Not. R. Astron. Soc. 000, 000–000 (0000) Printed 10 May 2019 (MN LATEX style ﬁle v2.2)

David O. Cook,1,2 Mansi M. Kasliwal,1 Angela Van Sistine,3 David L. Kaplan,3 Jessica S. Sutter,4 Thomas Kupfer,1 David L. Shupe,2 Russ R. Laher,2 Frank J. Masci,2 Daniel A. Dale,4 Branimir Sesar5 Patrick R. Brady,2 Lin Yan,4 Eran O. Ofek,6 David H. Reitze,1 Shrinivas R. Kulkarni1 1California Institute of Technology, 1200 East California Blvd, Pasadena, CA 91125, USA; [email protected] 2 IPAC/Caltech , 1200 E California Blvd, Pasadena, CA 91125, USA 3University of Wisconsin-Milwaukee, P.O. Box 413, Milwaukee, WI 53201, USA 4University of Wyoming, 1000 University Ave, Laramie, WY 82071, USA 5Max Planck Institute for Astronomy, Knigstuhl 17, D-69117 Heidelberg, Germany 6Weizmann Institute of Science, Rehovot, Israel

10 May 2019

ABSTRACT

We present the Census of the Local Universe (CLU) narrow-band survey to search for emission-line (Hα) galaxies. CLU-Hα has imaged ≈3π of the sky (26,470 deg2) with 4 narrow-band filters that probe a distance out to 200 Mpc. We have obtained spectroscopic follow-up for galaxy candidates in 14 preliminary fields (101.6 deg2) to characterize the limits and completeness of the survey. In these preliminary fields, CLU can identify emission lines down to an Hα flux limit of 10−14 erg s−1 cm−2 at 90% completeness, and recovers 83% (67%) of the Hα flux from catalogued galaxies in our search volume at the Σ=2.5 (Σ=5) color excess levels. The contamination from galaxies with no emission lines is 61% (12%) for Σ=2.5 (Σ=5). Also, in the regions of overlap between our preliminary fields and previous emission-line surveys, we recover the majority of the galaxies found in previous surveys and identify an additional ≈300 galaxies. In total, we find 90 galaxies with no previous distance information, several of which are interesting objects: 7 blue compact dwarfs, 1 green pea, and a Seyfert galaxy; we also identified a known planetary nebula. These objects show that the CLU- Hα survey can be a discovery machine for objects in our own Galaxy and extreme galaxies out to intermediate redshifts. However, the majority of the CLU-Hα galaxies identified in this work show properties consistent with normal star-forming galaxies. arXiv:1710.05016v2 [astro-ph.GA] 8 May 2019 CLU-Hα galaxies with new redshifts will be added to existing galaxy catalogs to focus the search for the electromagnetic counterpart to gravitational wave events. Key words: galaxies: dwarf – galaxies: irregular – Local Group – galaxies: spiral – galaxies: star formation

1 INTRODUCTION and blue compact dwarfs (BCDs). Many of these galaxies exhibit some of the most extreme properties still known in Large-area, blind surveys of emission-line or UV-excess the local Universe (low metallicity, high star formation rates, galaxies have led to important extra-galactic discoveries optical line ratios, etc.; Mickaelian 2014), and have facili- over the past few decades. The ﬁrst systematic search was tated a better understanding of star formation and galaxy the Markarian Survey (Markarian, Lipovetskii & Stepanian evolution. 1981) which in total scanned 15,200 deg2 from 1965-1980 Several studies have since searched smaller areas of down to an characteristic magnitude of 15.5 mag in the V- the sky to greater depths, such as the second Markar- band. Just over 1500 galaxies were discovered in this search ian Survey (Markarian & Stepanian 1983), University of including starburst galaxies, AGN, QSOs, Sefert galaxies, Michigan (UM) survey (MacAlpine, Smith & Lewis 1977a),

c 0000 RAS 2 D. Cook et al.

Case survey (Pesch & Sanduleak 1983), Kitt Peak Inter- resolution examinations of galaxies located in the local Uni- national Spectral Survey KISS (Salzer et al. 2000), the verse. As such, an all-sky, or nearly all-sky, survey designed Hamburg/SAO Survey (HSS; Ugryumov et al. 1999). Many to find a more complete sample of extreme and normal star- of these studies focused on finding samples of BCDs to forming galaxies in the local Universe would provide better investigate their relationship to primordial environments statistics of galaxy evolutionary trends and would sample due to their low metallicities (i.e., as low as 1/40 Z or more extreme environments from which to test star forma- 12+log(O/H)=7.1 for DDO68, SBS 0335052, and I Zw 18; tion theories. Thuan & Izotov 2005; Izotov, Thuan & Guseva 2005; Izotov In this paper we introduce such a survey: the Census of & Thuan 2007) and their high gas content (see Thuan 1991 the Local Universe (CLU) emission-line (Hα) galaxy survey. for a review). In the decades after these surveys, a handful CLU-Hα is a narrow-band survey of the entire northern sky of extremely low metallicity BCDs have been found (Izotov above a declination of −20 deg with the aim of constrain- et al. 2006; Izotov & Thuan 2007; Hirschauer et al. 2016; ing galaxy distances out to 200 Mpc. We have imaged ≈3π Izotov et al. 2018), but none with 12+log(O/H)<6.9 sug- sr of the sky as part of the Intermediate Palomar Tran- gesting a metallicity floor in the local Universe. However, sient Factory (iPTF; Law et al. 2009) with four contiguous larger area surveys that can select BCDs could fill out the narrow-band filters. Using the Hα emission line, CLU is de- extremely low metallicity end of the distribution and provide signed to find emission-line galaxies from redshift 0 to 0.047 constraints on star formation and galaxy evolution models. (≈200 Mpc). As a consequence, CLU will provide distance The next advancement in large area galaxy surveys was constraints and SFRs to normal star-forming galaxies with the Sloan Digital Sky Survey (SDSS; York et al. 2000). moderate EW emission lines and provide a more complete The SDSS spectroscopic galaxy survey obtained spectra of census of galaxies in the local Universe. In addition, CLU half a million extragalactic sources in ∼9400 deg2 using will also provide lists of extreme galaxies such as: BCDs in fiber-fed plates (Alam et al. 2015). These data directly led the local Universe, and green peas at intermediate redshifts to the discovery of the “Main Sequence” of star-forming whose redshifted [OIII] lines will also be detectable in our galaxies, where the current star-formation rate (SFR) of a survey. A more complete list of extreme galaxies in the local galaxy forms a tight relationship with its total stellar mass Universe across 3/4ths the sky will probe more extreme en- (Brinchmann et al. 2004; Daddi et al. 2007; Salim et al. vironments and put these galaxies in context of the galaxy 2007; Peng et al. 2010). This, now fundamental, relation- “Main Sequence.” ship led to further discoveries which found that the “Main Another astrophysical impact of the CLU survey is the Sequence” shows an invariant scatter but an increased nor- search for electromagnetic (EM) counterparts to gravita- malization at higher redshifts (Elbaz et al. 2007; Noeske tional waves (GW) now detectable from the Laser Inter- et al. 2007; Heinis et al. 2014). Thus, these studies have ferometer Gravitational Wave Observatory (LIGO; Abbott provided the means to quantify the evolution of galaxy prop- et al. 2016). Linking the new discovery tool medium of gravi- erties over time, and can provide insights into star formation tational waves to our understanding of electromagnetic radi- via galaxies whose properties fall off the “Main Sequence.” ation observable from our current telescopes will have signif- The SDSS survey also gave rise to the discovery of icant impacts on our understanding of the Universe. How- extreme emission-line galaxies at intermediate redshifts ever, the sky localization of LIGO GW events can range (0.11 ≤ z ≤ 0.36) known as “green peas.” These objects from 30–1000 square degrees on the sky making this search were identified based on their point source-like appearance a challenge. Targeted follow-up observations of likely host where the strong [OIII]λ4959, λ5007 doublet (EW≥100 A)˚ galaxies can narrow down the search area by a factor of 100 is located in the green-coded SDSS r-band filter giving them (Nissanke, Kasliwal & Georgieva 2013; Gehrels et al. 2016). their green colors (Lintott et al. 2008). In-depth studies of As such, the CLU narrow-band filters were designed to probe green peas have shown that these objects have compact out to the projected sensitivity distance of merging objects morphologies, high SFRs, and low metallicities (Cardamone thought to produce EM counterparts (i.e., neutron stars at et al. 2009; Amor´ın, Pérez-Montero & V´ılchez 2010; Izo- 200 Mpc; Aasi et al. 2015). tov, Guseva & Thuan 2011). Thus, these galaxies bare a In this paper, we introduce the CLU-Hα survey, pro- striking resemblance to high-redshift galaxies (Lyman-break vide the survey parameters (e.g., sky coverage, narrow-band galaxies and Lyα emitters) that are thought to contribute a filter properties, and survey limits), detail the galaxy can- significant fraction of ionizing photons during the epoch of didate selection criteria, show examples of extreme objects reionization (Finkelstein et al. 2012). In addition, the similar found in preliminary fields, and discuss science applications metallicities and abundance ratios of green peas and BCDs (e.g., extreme galaxies, star-forming galaxies, and EMGW suggest that these two are related objects (Izotov, Guseva counterpart searches) for this unique survey. & Thuan 2011), but this connection is currently under de- bate. Large area surveys that can select “green peas” and BCDs will provide better statistics to help determine if a 2 CLU-Hα SURVEY DESCRIPTION connection exists. The discovery of the galaxy “Main Sequence” has al- The CLU-Hα survey endeavors to discover as many Hα- lowed astronomers to quantify galaxy properties over time emitting galaxies as possible out to 200 Mpc (z=0.047) and and the low scatter in this relationship has provided a refer- provide a relatively well constrained redshift range for each ence point from which to put extreme galaxies into context. emission-line galaxy. In addition, the Hα fluxes measured for However, a better understanding of the physical processes all galaxies (both previously known and newly discovered in responsible for the “Main Sequence” scatter and the extreme the local Universe) will provide a recent SFR for each galaxy deviations will largely come from high sensitivity and spatial (i.e., the integrated SFR over the past ∼10 Myr; e.g., Ken-

c 0000 RAS, MNRAS 000, 000–000 CLU I: Galaxy Catalogs from Preliminary Fields 3

Narrowband Hα Filter Properties

Filter Filter Filter Redshift name λ ∆λ range (A)˚ (A)˚ (#)

Hα1 6584.2 76.1 -0.0026 < z < 0.0090 Hα2 6663.7 77.9 0.0094 < z < 0.0213 Hα3 6730.9 90.1 0.0187 < z < 0.0324 Hα4 6822.1 92.1 0.0325 < z < 0.0471

Table 1. The properties of the CLU narrowband filters, where the columns present the filter names, central wavelength, FWHM, and redshift range, from left-to-right. The first filter (Hα1) is centered on rest-frame Hα while the last filters FWHM extends to 200 Mpc.

first two filters (Hα1 and Hα2) make up the first filter pair The measured filter transmission profiles of the 4 Figure 1. and the 3rd and 4th filters (Hα3 and Hα4) make up the CLU-Hα narrow-band filters where the blue, cyan, green, and red curves represent Hα1, Hα2, Hα3, and Hα4 filters, respec- second filter pair. We note that the small overlap between tively. The horizontal dotted and dashed lines represent the CCD the filter pairs can result in decreased efficiency of emission quantum efficiency and the atmospheric transmission at Palomar line detection for emission lines whose wavelengths fall in Observatory, respectively. this gap; however, the overlap is only 10 A˚ and is not likely to affect a large number of sources. nicutt 1998; Murphy et al. 2011; Kennicutt & Evans 2012). Thus CLU-Hα will not only yield the positions and well 2.2 Observational Strategy constrained distances for new galaxies in the local Universe, but will also provide nearly uniform Hα fluxes and SFRs for Observations were taken on the Oschin 48 inch telescope 2 both newly discovered and previously known galaxies. at the Palomar Observatory with a 7.92 deg mosaic imager composed of 12 CCD detectors (or chips) and a 100 per pixel image scale. However, one of the chips (chip #3) is 2.1 Narrow-band Filters non-functional resulting in an 11-chip imager with a field of view 7.26 deg2 (for details see, Law et al. 2009; Rau CLU-Hα searches for emission-line galaxies using 4 et al. 2009). The survey uses an observational strategy of wavelength-adjacent, narrow-band filters with a combined 3 spatially-staggered, overlapping grids that are limited to a ˚ wavelength range of 6525−6878 A. Each emission-line galaxy declination of greater than −20◦, where each grid contains can be identified via a flux excess in one filter (the “On” fil- N= 3626 iPTF fields (26, 470 deg2 of the sky). Observations ter) signifying the presence of an emission line compared to using the Hα1 and Hα2 filters cover the entire sky above a filter that covers only the adjacent continuum (the “Off” −20◦ declination. Observations using the Hα3 and Hα4 fil- filter). Thus, CLU-Hα provides a distance constrained by ters also cover the sky above −20◦ declination, but avoid the width of our narrow-band filters. ◦ the Galactic plane (|b| & 3 ). A single 60 second exposure is The transmission curves for our narrow-band filters are taken for each field in each grid resulting in three images for shown in Figure 1, where the solid lines represent the trans- the majority of positions on the sky. The multiple images mission from the manufacturer, the dotted line represents facilitate cosmic ray rejection, filling in chip gaps and the the CCD quantum efficiency of the camera and the dash- non-functional chip, and deeper final co-added images. dotted line represent the atmospheric transmission at Palo- Data acquisition ended in March 2017 (due to decom- mar (Ofek et al. 2012). missioning of the iPTF camera) with 98.3% of fields ob- The filter widths and central wavelengths are calculated served in Hα1 and Hα2 and 91.3% observed in Hα3 and via: Hα4. However, the second filter pair is 95% completed at a ◦ Galactic latitude limit of three degrees (95% at |b| & 3 ) and Z ◦ ◦ ∆λ ≡ T (λ)dλ, (1) 99% completed at latitudes above 11 (99% at |b| & 11 ). Figures 2 and 3 present the sky coverage maps in equatorial coordinates (where 0◦ Right Ascension is represented R λT (λ)dλ as the center, vertical red line) of the CLU-Hα survey for λ¯ ≡ , (2) the first filter pair (Hα1 and Hα2) and the second filter R T (λ)dλ pair (Hα3 and Hα4), respectively. The colors for each iPTF where T is final transmittance of the filter curves with the pointing box in these figures represent the number of obser- peak normalized to unity. Table 1 presents the properties of vations, where lighter/brighter colors indicate more obser- our 4 narrow-band filters. vations. The Galactic plane is easily identified in Figure 3 The first filter is centered near the z=0 Hα emission as a dark strip of fields with little to no observations in the line (λ = 6563 A)˚ and the wavelength range of the last filter second filter pair. The majority of the fields within our tar- extends to z=0.047 (i.e., ∼200 Mpc) Hα emission line. The geted sky coverage have three observations. In addition to

c 0000 RAS, MNRAS 000, 000–000 4 D. Cook et al. our survey, there were multiple iPTF projects that observed (CLU−PS1) and g − i (PS1-PS1) colors, where both CLU- specific regions of the sky using the CLU narrow-band fil- Hα and PS1 magnitudes are PSF fitted magnitudes. ters and can provide even deeper Hα imaging for these fields. Figure 4 graphically presents our calibration method, The locations that these programs targeted are apparent in where the grey dots represent all CLU-Hα sources matched Figures 2 and 3 as the brightest regions with ∼100-200 ob- to PS1 sources, the blue dots represent matched “good” servations. stars that meet our criteria, the dashed line represents the Although the ultimate goal of the CLU-Hα survey is zeropoint, and the solid-cyan line represents the fit to “good” to coadd the images from the 3 staggered, overlapping grid stars. The error in the fit is used as the zeropoint error and is patterns, the preliminary analysis here uses the single expo- added in quadrature to the photometric error of each source. sures from only one of these grids in 14 fields (see §3.1). Typical errors in the zeropoints are ∼ 0.05 mag. The fitted line in Figure 4 is used to perform a color correction to the magnitdues of the Hα sources. 2.3 Data Reduction and Source Catalogs Due to the large field of view of the iPTF instrument, Data reduction and source extraction are carried out in an variations in the sensitivity across the imaging array exist af- automated pipeline built by the Infrared Processing and ter the images are processed with the IPAC image reduction Analysis Center (IPAC)1 specifically for the iPTF survey. pipeline and need to be taken into account when deriving The full description of this pipeline can be found in Laher the final calibrated magnitudes of extracted sources. This et al. (2014), but we provide a brief overview here. variation can be quantified in the zero-point of stars located The IPAC reduction pipeline consists of both “off-the- across the entire field of view. The procedure for deriving a shelf” and custom software which have been extensively sensitivity variation map is described in detail by Ofek et al. tested on millions of images from iPTF broadband images. (2012), but we provide a brief overview here. Using a mini- After each night of data is acquired, the IPAC pipeline per- mum of 200 images for each chip and filter image we mea- forms a bias subtraction and applies a flat-field correction. sure the zero-points for all stars that meet our calibration The astrometric solution for each processed image is com- star criteria (see paragraph above), then combine them into puted via SCAMP (Bertin 2006) on one of three stellar cata- bins of 32 pixels on a side. The median zero-point residual logs: SDSS-DR7 (Abazajian et al. 2009), UCAC3 (Zacharias for each bin, normalized by the median zero-point across the et al. 2010), or USNO-B1 (Monet et al. 2003). entire image, are used as the zero-point variation correction. After data reduction is completed, Source Extractor This correction is applied to the final calibrated magnitude (Bertin & Arnouts 1996) is used to generate a source catalog for each source given the position on each chip and the fil- for every chip and filter image in each field. The fluxes are re- ter used in the observation. The zero-point variation across ported using many aperture definitions; however, we utilize a single chip has a typical range of 0.01-0.05 mag and the the fluxes from aperture photometry at 5 pixels (∼ 500) in di- standard deviation of the zeropoint variation is ∼0.02 mag ameter for galaxy candidate selection (see §3) and the point for all filters. spread function (PSF) fitted fluxes of stars for calibration purposes (see below). The 5 pixel aperture is used to select narrow-band excess sources since the Hα colors showed smaller scatter in this aperture size compared to larger 8 2.5 Ancillary Data and 10 pixel diameter apertures. The median and standard We supplement our CLU-Hα fluxes with information from 00 deviation 5σ detection limits of a point source with a 5 the SDSS DR12 (Alam et al. 2015), GALEX all sky (Martin diameter aperture for each chip image in all 14 preliminary et al. 2005), and WISE all sky surveys (Wright et al. 2010) in fields is 18.6 ± 0.30, 18.7 ± 0.28, 18.8 ± 0.28, 18.8 ± 0.25 AB addition to PS1 PSF and Kron magnitudes. We utilize the mag for filters Hα1, Hα2, Hα3, and Hα4, respectively. model ugriz magnitudes from SDSS DR12. In addition, we cross-match the CLU-Hα sources against entries in the SDSS ‘galSpecLine’ table and extract the redshift, Hα line flux 2.4 Calibration (‘h alpha flux’), and equivalent width (EW; ‘h alpha eqw’). Calibration is carried out for each chip and filter image com- We also extract FUV and NUV kron fluxes from the GALEX bination, where Pan-STARRS DR1 (PS1; Chambers et al. all-sky imaging survey (AIS; Bianchi, Conti & Shiao 2014), 2016) stars (N> 100) are matched with CLU-Hα sources. and the instrumental profile-fit photometry of the first and We match all CLU-Hα sources with the following criteria: fourth WISE bands (‘w1mpro’ and ‘w4mpro’). These ancil- 1) are relatively isolated (i.e., no source within 1000), 2) have lary data are used to cull contaminants and measure several a photometric error less than 0.1 mag, and 3) have no Source physical properties of galaxies. Extractor photometry flags to PS1 stars with the following criteria: 1) a g − i color between 0 and 1.7 mag, 2) a photometric error less than 0.1 mag, 3) a PSF minus Kron magnitude in the i-band of less than 0.05 mag to select stars.2 The photometric zeropoints and color correction terms are 3 CANDIDATE SELECTION determined by fitting a linear relationship between Hα − r In this section, we describe our galaxy candidate selection methods. We have chosen to test these methods in 14 prelim- 1 http://www.ipac.caltech.edu/ inary fields where we have performed a spectroscopic follow- 2 https://confluence.stsci.edu/display/PANSTARRS/How+to+ up campaign. In addition, we quantify the success rates of separate+stars+and+galaxies our selection method and the limits of the survey.

c 0000 RAS, MNRAS 000, 000–000 CLU I: Galaxy Catalogs from Preliminary Fields 5

Figure 2. The sky coverage map in equatorial coordinates for the first narrow-band pair of filters (Hα1 and Hα2) in our survey, where the red vertical line represents a RA=0◦. The Individual colored boxes represent one pointing with a 7.26 deg2 field of view. The color bar indicates how many observations have been taken for each pointing

Figure 3. The sky coverage map in equatorial coordinates similar to Figure 2 but for the second narrow-band pair of ﬁlters (Hα3 and Hα4) in our survey. The dark contiguous sections represents the Galactic plane which is avoided in the second ﬁlter pair.

3.1 Preliminary Fields troscopic follow-up from Palomar Observatory. The basic properties of the 14 preliminary ﬁelds are listed in Table 4.

The preliminary fields were chosen based on the following In addition, we have chosen to include one field which criteria: 1) the CLU-Hα images must contain Hα filter pairs contains a galaxy cluster. The field labeled “p3967” in Ta- taken on the same night and have observations in all four ble 4 is spatially coincident with the Coma cluster whose filters to ensure a complete analysis, 2) the fields must have redshift falls in the wavelength range of our third narrow- SDSS coverage to provide a list of galaxies with known red- band filter (Hα3). We include this field to robustly test the shifts, and 3) a declination close to 30◦ to facilitate spec- limits of our selection methods as this cluster contains hun-

c 0000 RAS, MNRAS 000, 000–000 6 D. Cook et al.

where con,off are the counts for the “On,Off” filters and δ is the sky fluctuations in both images combined in quadrature. The combined photometric uncertainty is expressed as:

q 2 2 2 δ = πr (σon + σoﬀ ); (4)

where σon/off represent the median sky count fluctuations in 100 sky regions for the “On” and “Off” images and r is the radius of the photometric aperture. This criteria is similar to a standard signal-to-noise selection and can be expressed in terms of measured colors and magnitudes via:

−0.4(ZP−mon) moﬀ − mon = −2.5log(1 − Σδ10 ) (5)

where mon/off are the calibrated magnitudes in the “On” and “Off” filters and ZP is the photometric zero-point for Figure 4. The Hα-r versus g −i color plot which is used to define the “On” filter. our photometric calibration (i.e., zeropoint and color correction). The relationship between Σ, color, and magnitude is The blue filled points are all stars selected for calibration, the represented as curved lines in a color-magnitude diagram horizontal blue line represents the zeropoint, and the solid cyan and are graphically presented in Figures 5−8 for each Hα fil- line is the fit to calibrations stars and is used for color corrections. ter in one field (“p3967”). While all four of these figures are similar in form, we enlarge the first figure for clarity. In each of these figures the black points represent all sources in the dreds of galaxies whose Hα EWs span a large range from field, the red dots represent galaxies with a spectroscopic ellipticals with no emission lines to star bursts with strong redshift (from SDSS) greater than the redshift coverage of emission lines (Mahajan, Haines & Raychaudhury 2010). our filter set (z > 0.047), the green X’s represent galaxies However, the majority of the preliminary fields occupy rel- with a spectroscopic redshift (from SDSS) that fall in the atively sparse sections of the sky (i.e., outside the Galactic wavelength range of the appropriate filter, and the green plane and not coincident with any galaxy clusters). Each circles represent the galaxy candidates selected in each fil- 2 pointing in our Hα survey covers 7.26 deg on the sky; thus ter. 2 our 14 preliminary fields cover a total of 101.6 deg . Inspection of Figures 5−8 also shows a non-zero color scatter even at brighter magnitudes. Thus, we impose a second condition based on the standard deviation of colors for 3.2 Candidate Selection Criteria bright continuum sources (i.e., stars) with magnitudes be- The identification of emission-line candidates follows the tween 12 and 15 mag. The second criteria is represented as general methods of previous emission-line galaxy surveys horizontal lines in Figures 5−8, and represent the minimum using narrow-band filters (e.g., Bunker et al. 1995; Pascual color below which we cannot accurately infer the presence of et al. 2001; Fujita et al. 2003; Geach et al. 2008; Sobral et al. an emission line. Furthermore, since the EW of an emission 2009; Ly et al. 2010; Lee et al. 2012; Sobral et al. 2012; Stroe line is simply the ratio of the line flux and the continuum & Sobral 2015). The photometry used for the selection pro- flux density, we can calculate an EW limit based on the cess are taken from the Source Extractor catalogs. We use “On-Off” color scatter in bright continuum sources. Follow- the 5 pixel diameter aperture photometry for all selection ing the prescription of (Stroe & Sobral 2015, see Equation criteria. In addition, we require a candidate to have a 5σ 7), the EW can be calculated via: detection in the “On” band and set non detections to the 5σ upper limit (see Table 4). −0.4∆m EW = ∆λon 10 − 1 , (6) Emission-line candidates are initially selected based on the significance of the excess in “On-Off” color quantified where ∆λon is the FWHM of the “On” filter and ∆m is the by the parameter Σ, where the flux is greater in one filter “On-Off” color. The EW cuts based on the “On-Off” colors (the “On” filter) due to the presence of an emission line for each filter in Figures 5−8 are labeled on the horizontal compared to the corresponding continuum filter (the “Off” lines. filter). Each filter pair (Hα1/Hα2 and Hα3/Hα4) provides Both of the selection criteria adopted here (see, Bunker both the “On” and “Off” photometry. For example, Hα2 is et al. 1995; Fukugita et al. 2007; Ly et al. 2010; Lee et al. used as the “Off” filter for Hα1 selection, while Hα1 is used 2012; Sobral et al. 2009; Sobral et al. 2012; Stroe & So- as the “Off” filter for Hα2 selection. bral 2015) represent the number of standard deviations away The significance of the narrow-band colors (Σ) is defined from the expected random scatter and can be combined into as the number of standard deviations between the color ex- a single Σ value. For each source, the number of standard cess (“On-Off”) in counts and the random scatter of counts deviations is computed above each criteria (bright stars and expected for a source with zero color (Bunker et al. 1995). the noise in the images), and the final Σ is defined as the This can be expressed as: smaller of the two. For example, the Σ of a bright source c − c (e.g., mon . 17 mag) will be calculated relative to the scat- Σ = on off , (3) δ ter in continuum sources, and the Σ of a faint source (e.g.,

c 0000 RAS, MNRAS 000, 000–000 CLU I: Galaxy Catalogs from Preliminary Fields 7

Figure 5. The Hα color-magnitude diagram of objects in the field labeled “p3967” , where the y-axis represents the “On-Off” color and the x-axis is the “On” magnitude. The black points represent all CLU-Hα sources, the red dots represent contaminant galaxies, the green X’s represent known galaxies with a redshift that fall in the wavelength range of the Hα1 filter, and the green circles represent the galaxy candidates selected. The dotted and dashed lines represent the Σ cuts of 2.5 and 5, respectively. The “On-Off” color defined by the standard deviation of bright stars is converted into an Hα EW limit and is labeled for the Σ = 5 cut (horizontal dashed line) with a value of 16.3 A.˚ The curved lines are derived from a signal-to-noise criteria where the “On-Off” color scatter increases towards fainter magnitudes. Sources with colors that exceed the Σ thresholds and are not selected as candidates were identified as contaminants (e.g., stars; see §3.3).

mon & 17 mag) will be calculated relative the signal-to-noise curve. We utilize two Σ cuts to define the extremes of our candidate selection methods. The first is a cut of Σ = 2.5 which will contain the majority of all star-forming galaxies but will contain a relatively higher contamination from galaxies with no emission lines in our narrow-band filters. The second is a cut of Σ = 5 which will contain a reduced fraction of all star-forming galaxies but will contain a low contamination fraction (see §3.5). The median EW cut and standard deviation for all filters in the 14 preliminary fields is 7.3 ± 1.1 A˚ and 15.2 ± 2.4 A˚ for Σ = 2.5 and Σ = 5, respectively. We provide an electronic version of the CLU candidates with Σ values above 2.5. In the future, we will generate source catalogs in larger areas of the sky where Σ values will be calculated for each source and users can apply their own cuts based upon their science requirements. For a bet- Figure 6. The Hα color-magnitude diagram used for emission- ter understanding of the success and contamination rates at line sources in the Hα2 filter similar to that in Figure 5. The different Σ values see §3.5. “On-Off” color defined by the standard deviation of bright stars is converted into an Hα EW limit and is labeled for the Σ = 5 cut (horizontal dashed line) with a value of 16.8 A.˚ 3.3 Contaminant Removal The two color excess cuts introduced in the last section will effectively select any source that has a significant Hα “On- Off” color. However, the resulting galaxy candidates can still

c 0000 RAS, MNRAS 000, 000–000 8 D. Cook et al.

FWHM∼ 200, we estimate that galaxies larger than 2 kpc at 200 Mpc will be extended in our data. Comparison to a statistically complete sample of star-forming galaxies in the local Universe where the sample is dominated by dwarf galaxies (LVL; Kennicutt et al. 2008; Dale et al. 2009; Lee et al. 2011; Cook et al. 2014a) reveals that the semi-major axis histogram peaks between 2 − 4 kpc suggesting that we will likely detect the majority of galaxies even at our furthest distance by requiring our galaxy candidates to be extended. Similar results are found for the size of Hα disks in galaxies out to z=0.1 (Dale et al. 1999). We exclude point sources based upon the recommended PS1 star/galaxy separation using PSF minus Kron i-band magnitudes. Unfortunately, the PS1 star-galaxy classifications of- ten mislabel saturated stars as galaxies at r-band magnitudes brighter than 12-14 mag3 leaving ∼300 candidates that are clearly stars. We can remove ∼60% of these con- Figure 7. The Hα color-magnitude diagram used for emission- taminants via a bright Hα magnitude cut of 12 mag given line sources in the Hα3 filter similar to that in Figure 5. The the brightest quoted saturation magnitude in PS1. We note overabundance of galaxies for this filter is due to the Coma cluster that the brightest confirmed galaxy in the preliminary fields which is dominated by early-type galaxies with little on-going star is ∼13.4 mag. The remaining contaminating stars are re- formation. The lack of star formation in these galaxies is evident moved via visual classification. from the lack of significant Hα “On-Off” colors. The “On-Off” The removal of cosmic rays and chip defects is achieved color defined by the standard deviation of bright stars is converted by requiring a source to be spatially coincident with a PS1 into an Hα EW limit and is labeled for the Σ = 5 cut (horizontal source. Since the PS1 detection limits (e.g., r ∼ 23.2 mag) dashed line) with a value of 19.8 A.˚ are deeper than the CLU-Hα single-image exposures, any cosmic ray or chip defect with no spatial overlap with a PS1 source will be easy to flag and remove. Visual inspection of sources removed from our analysis via this method show that no real sources have been removed. Furthermore, visual inspection of our galaxy candidates show no contamination from cosmic rays, but do show a small percentage of contamination from chip-column defects that randomly coincide with a PS1 source; these contaminants are easily removed via visual inspection. We note that the removal of cosmic rays and chip defects will be greatly reduced or unnecessary for future analyses using stacked images. The final sources of contamination are high-redshift galaxies whose emission or absorption lines have been shifted into the wavelength range of our filters. It is not possible to distinguish between strong emission lines redshifted into the wavelength range of our filters and Hα emission at lower redshift. We note that there is a dearth of strong emission lines blue-ward of our filters, where the closest lines are the [OIII]λ4959, λ5007 doublet near λ =5000 A˚ (z∼0.3). How- Figure 8. The Hα color-magnitude diagram used for emission- ever, galaxies with extreme [OIII]λ4959, λ5007 emission at line sources in the Hα4 filter similar to that in Figure 5. The z∼0.3 are likely to be interesting objects (i.e., green peas) “On-Off” color defined by the standard deviation of bright stars and can be used for studies of galaxy evolution and star for- is converted into an Hα EW limit and is labeled for the Σ = 5 mation. In fact, we find a newly discovered z∼0.3 green pea ˚ cut (horizontal dashed line) with a value of 19.8 A. contaminant in the preliminary fields (see §5.1) and future CLU efforts will be aimed at finding and studying more of these objects. be contaminated by continuum sources with steep blue or Absorption lines in elliptical or red-sequence galaxies at red continuum slopes, high-redshift galaxies with an emis- intermediate redshifts are another source of contamination. sion or absorption line whose wavelength has been redshifted Strong NaD absorption at redshifts of 0.11–0.17 can cause a into the wavelength range spanned by one of our filters, and lower flux in an Hα filter relative to the flux in its pair filter by cosmic rays or chip defects (e.g., hot pixels, column de- resulting in a significant “On-Off” color. Since red sequence fects, etc.). galaxies tend to be redder than star-forming galaxies (Salim Point sources with a steep blue or red continuum can et al. 2007), we can remove these contaminants via simple be reduced by requiring that our candidates be spatially extended. As our survey is only sensitive to a distance of 200 Mpc and our angular resolution is limited to a 3 https://panstarrs.stsci.edu/

c 0000 RAS, MNRAS 000, 000–000 CLU I: Galaxy Catalogs from Preliminary Fields 9

the spectroscopic measurements. In Figure 10 we plot the photometric Hα line flux versus the spectroscopic line flux in the left panel. The photometric Hα line flux is measured via:

−1 −2 Fline(erg s cm ) = ∆λOn(fOn − fOﬀ ), (7)

where ∆λOn is the FWHM of the “On” filter, fOn is the flux density of the “On” filter, and fOff is the flux density of the “Off” filter. The flux densities of the “On” and “Off” filters are calculated via:

−1 c −1 −2˚ −0.4(mon,off +ZptAB) fon,off (erg s cm A ) = 2 10 , λon,off (8) where c is the speed of light, ZptAB = 48.59, and λon,off are the central wavelengths of the “On” and “Off” filters. Figure 9. The g − r versus NUV-r, color-color plot of all known There is reasonable agreement for the majority of galax- galaxies in the preliminary fields, where the blue squares represent ies above a few×10−15erg s−1 cm−2 for line fluxes derived galaxies in our target volume with an Hα EW greater than the Σ = 2.5 limit (EW≥7.5 A),˚ the green pluses are all galaxies in from our Hα photometry and those derived from spec- our volume with small Hα EWs (EW<7.5), and the red points are troscopy. However, there is a small offset (a few tenths of galaxy candidates outside of our target volume (z>0.047). A g −r a dex) at lower line fluxes where the photometrically de- optical color cut removes contaminant galaxies while retaining the rived fluxes are systematically higher than the spectroscopic majority of galaxies in our volume with an EW greater than our fluxes. This offset is likely due to a mismatch in the aper- selection limits (EW≥7.5A).˚ tures used to derive the line fluxes, and from contaminating flux from [NII] in the Hα imaging. We use a 500 aperture in the Hα imaging, while the spectroscopic line fluxes are 00 00 optical color cuts. Figure 9 shows the g − r versus NUV − derived from either a 3 fiber from SDSS or a 2 slit from r color-color diagram for galaxies in our preliminary fields Palomar observations. In addition, the [NII]/Hα ratio for with low and high Hα EWs as measured via spectra from star-forming galaxies in the local volume can range from SDSS and CLU-Hα follow-up. We find that a g − r color cut 0.05 to 0.5 with an average of 0.2 (Kennicutt et al. 2008). of g−r greater than to 0.9 mag removes 60% of red sequence These ratios translate into an average and max offset in Fig- galaxies at z > 0.047 while removing only 4 galaxies with ure 10 of 0.1 and 0.3 dex, respectively, and is likely the major moderate EW and a redshift less than 0.047. source of the offsets seen in Figure 10. We also note that the strength of the [NII] contamination depends on the redshift of the galaxy where the [NII] lines can be shifted out of the 3.4 Spectroscopic Follow-up “On” filter if the Hα line is near the filter edge. We have obtained spectra for 334 candidates in the preliminary fields with no redshift information (except obvious 3.5 Comparison to Known Galaxies stars, cosmic rays, and chip defects) with a Σ value above 2.5. The spectra were taken with the 200 inch Hale tele- In this section, we evaluate our selection criteria by cross- scope atop Palomar Mountain on multiple nights over 2016 matching our galaxy candidates to currently cataloged and 2017 using the Double-Beam Spectrograph instrument galaxies with spectroscopic redshifts available from NED, (DBSP; Oke & Gunn 1982) and on the 2.3 meter Wyoming SDSS, or our CLU-Hα followup in the preliminary fields. Infra-Red Observatory (WIRO) with the Longslit spectro- Since our selection methods rely on the presence of a mod- graph. The DBSP data were reduced using the PyRAF-based erately strong Hα emission line, we expect that we should pipeline4 of Bellm & Sesar (2016) and the WIRO data were recover a large fraction of the Hα flux for galaxies in these reduced with standard IRAF procedures. fields, but recover a relatively low fraction by number of We took spectra of 334 galaxies where we confirm these galaxies due to the numerous elliptical galaxies in these that 124 galaxies were indeed galaxies with redshifts less fields (i.e., in the Coma cluster). than 0.047, while the remainder were higher-redshift con- First, we examine the composition of galaxy catalogs taminants. In addition to finding new galaxies in the tar- produced when using different Σ cuts: 2.5, 3, 4, 5, 6, and 7. get volume, we also found 2 new galaxies (a Seyfert 1 Table 2 shows the total number of galaxy candidates, con- galaxy and a green pea) at intermediate redshift via strong taminant galaxies (i.e., those with no emission line found [OIII]λ4959, λ5007 emission lines redshifted into our filters in our narrow-band filters), galaxies found in the volume of (see §5.1). our survey (z<0.047), galaxies with an emission line found in In the rest of this section we compare the photometric the correct filter (i.e., an emission line has been confirmed in fluxes measured in our survey to with those derived from the filter identified by our narrow-band colors), and galaxies with new distance measurements (i.e., with no previous distance measurements) now confirmed in the local Universe 4 https://github.com/ebellm/pyraf-dbsp by our survey in the correct filter.

c 0000 RAS, MNRAS 000, 000–000 10 D. Cook et al.

Figure 10. Left Panel) The photometrically-derived Hα line flux versus the spectroscopically-derived Hα line flux in the left panel. We find that the majority of the galaxies show good agreement for Hα line fluxes between those derived from imaging and spectroscopy above a few×10−15erg s−1 cm−2. Right Panel) The histogram of the Σ = 2.5 and Σ = 5 CLU-Hα galaxy samples.

We find that the percentage of contaminant galaxies we recover 82.5% and 67.8% for Σ cuts of 2.5 and 5, re- falls significantly above a Σ cut of 3, and that the percentage spectively. If we limit the sample of known galaxies to those of galaxies with an emission line found in the correct filter is with a moderate Hα EW (> 7.5 A),˚ we recover 86.7% and ≈ 90% above a Σ cut of 5. Thus, we conclude that a Σ cut 72.0% by Hα flux for Σ cuts of 2.5 and 5, respectively. The above 5 produces a high fidelity galaxy catalog. However, a success rates of other Σ samples can be found in column 5 Σ cut above 2.5 does produce a catalog with a higher number of Table 3). of galaxies with an emission line found in the correct filter, Next, we examine why galaxies with an EW greater but with a high contamination percentage. In this and future than 7.5 A˚ are not selected as candidates (i.e., our false- publications we will release galaxy candidates with Σ cuts negative rate). The top panel of Figure 11 shows the above 2.5, and as a consequence we focus the text in the Hα “On” magnitude histogram for these galaxies with Σ >= rest of this section and the paper on the galaxy catalogs 2.5 and a Σ < 2.5, and find that the galaxies missed in our from two extreme cuts (Σ > 2.5 and Σ > 5) but provide the survey that should be selected tend to be fainter. In these statistics for all Σ cuts in the tables. fainter galaxies, the significance of their narrow-band colors is reduced by the increased photometric noise and thus re- Next, we quantify the success rates of our selection quire a stronger EW line to be selected by our methods (See methods both by number and by Hα flux in Table 3 via the bottom panel of Figure 11). We note that the photomet- a comparison to galaxies with known redshifts in our sur- ric errors will be reduced in the final stacked CLU-Hα images vey volume and spatially located in our preliminary fields. resulting in higher success rates in future analyses. The simplest comparison we can make is by number where we successfully recover 25.1% and 12.9% for Σ cuts of 2.5 and 5, respectively (column 2). However, There are 1226 known galaxies in our preliminary fields with known red- 3.6 Survey Limits shifts, where 722 have an Hα EW lower than 7.5 A(˚ ≈60%). In this section we explore the limits of our selection criteria The majority of these galaxies are located in a single point- and quantify them by Hα flux. We find that our survey is ing that is coincident with the Coma cluster (i.e., ‘p3967’). 90% complete at an Hα flux of 1 × 10−14 erg s−1 cm−2 and Thus a low success rate by number is not surprising given 4×10−14 erg s−1 cm−2 for the Σ=2.5 and Σ=5, respectively. the large fraction of elliptical galaxies in our preliminary To quantify the limits of this survey, we present the fields. If we limit the sample to galaxies with an Hα EW Hα flux histograms for our confirmed galaxy candidates and larger than the limits for the Σ = 2.5, we find that we re- all galaxies in the preliminary fields with spectroscopic in- cover 54.5% and 29.8% by number for Σ cuts of 2.5 and 5, formation from either SDSS or our own follow-up. In the respectively (column 4). top panel of Figure 12, the unfilled, red-filled, and blue- Since our selection methods are based on the strength filled histograms represents all galaxies with spectroscopy of the Hα emission line, it is more appropriate to quantify and our CLU-Hα galaxies with Σ > 2.5 and with Σ > 5. our success rates via Hα flux. Column 3 of Table 3 shows the The completeness (defined as the fraction of all galaxies re- fraction of Hα flux captured by each of the Σ cuts, where covered in our survey in each line flux bin) is illustrated

c 0000 RAS, MNRAS 000, 000–000 CLU I: Galaxy Catalogs from Preliminary Fields 11

Color Excess (Σ) Statistics

Σ N Total N Contaminant N In N in Correct N New Distances Candidates Galaxies Volume Hα Filter & in Correct Hα Filter (#) (#) (#,% of candidates) (#) (#,% of candidates) (#)

≥2.5 663 405 (61.1%) 339 258 (38.9%) 90 ≥3.0 399 174 (43.6%) 265 225 (56.4%) 78 ≥4.0 218 49 (22.5%) 180 169 (77.5%) 56 ≥5.0 147 18 (12.2%) 135 129 (87.8%) 41 ≥6.0 106 7 ( 6.6%) 100 99 (93.4%) 32 ≥7.0 86 3 ( 3.5%) 83 83 (96.5%) 23

Table 2. The composition of our galaxy candidates when using different Σ cuts, where the columns are from left-to-right: the minimum Σ for the sample, the total number of all candidates, the number of galaxies with no emission-lines in our narrow-band filters, the number of galaxies found in the survey volume (z<0.047), the number of galaxies whose Hα emission-line is located in the narrowband filter where the galaxy was identified, and the number of galaxies with no previous distance information found in CLU-Hα.

Known Galaxies

Sample N Sum Hα Flux N Sum Hα Flux (#) (erg/s/cm2) (#) (erg/s/cm2) (EW>7.5 A)˚ (EW>7.5 A)˚ (N,% of N) (Flux,% of Flux) (N,% of N) (Flux,% of Flux)

All 1173 4.05e-12 477 3.80e-12

Σ ≥ 2.5 295 (25.1%) 3.34e-12 (82.5%) 260 (54.5%) 3.30e-12 (86.7%) ≥3.0 259 (22.1%) 3.23e-12 (80.0%) 232 (48.6%) 3.20e-12 (84.1%) ≥4.0 192 (16.4%) 2.93e-12 (72.5%) 179 (37.5%) 2.92e-12 (76.9%) ≥5.0 151 (12.9%) 2.74e-12 (67.8%) 142 (29.8%) 2.74e-12 (72.0%) ≥6.0 120 (10.2%) 2.55e-12 (63.1%) 112 (23.5%) 2.55e-12 (67.1%) ≥7.0 102 ( 8.7%) 2.34e-12 (57.7%) 96 (20.1%) 2.34e-12 (61.4%)

Table 3. The success rates for galaxy catalogs generated with different Σ values. The first column gives the Σ values used to generate different galaxy samples, and the following columns give the success rates by number and Hα flux. The vertical lines are used to separate the success rates given no EW cut and an EW>7.5 A.˚ We find that our Σ = 2.5 sample recovers ∼25% by number and ∼82% by Hα flux compared to all galaxies, and that we recover a larger fraction by number and Hα flux when comparing to galaxies with an EW greater our selection limits. in the bottom panel of Figure 12. We find that we recover emission lines. We note that the ”X” symbols in Figure 13 −14 90% and 50% of known galaxies at Fline = 1 × 10 and represent galaxies with no previous distance information in −15 −1 −2 −14 3 × 10 erg s cm for Σ > 2.5 and Fline = 4 × 10 NED or SDSS. and 1 × 10−14 erg s−1 cm−2 for Σ > 5. Next, we explore the redshift-magnitude distribution of the two CLU-Hα galaxy catalogs in Figure 13 to illustrate 4 RESULTS the resulting CLU-Hα galaxy catalogs and their magnitudes. In this section we provide a description of the CLU- The top panel presents the full redshift range of our galaxy Hα galaxies found in the preliminary fields and highlight candidates while the bottom panel is a zoom-in of the local two sub-samples composed of: 1) galaxy candidates with Universe (z<0.05) In addition, the dashed- and solid-curved lower color significance (Σ ≥2.5) that will contain the ma- lines represent the relationship between an apparent magni- jority of target galaxies (i.e., those in the survey volume tude limit of 17.8 (i.e., the limit of the SDSS spectroscopic with Hα emission lines) but with high contamination; 2) galaxy survey) and 19 mag, respectively, with redshift. galaxy candidates with higher color significance (Σ ≥5) that The density of all galaxies with spectroscopic redshifts contain a reduced fraction of target galaxies but with low drops off significantly below the r-band magnitude limit of contamination. First, we present examples of galaxies whose the SDSS spectroscopic galaxy survey (r=17.8 mag). In ad- redshifts have been measured for the first time in this survey. dition, the density of CLU-Hα galaxies also significantly Then, we examine the observable properties of the galaxies decreases near that of SDSS suggesting that our selection in our CLU-Hα catalog and derive their physical proper- methods result in an effective magnitude limit similar to ties. Finally, we compare the CLU-Hα galaxies found in the the SDSS limit. Our methods do not select bright galaxies preliminary fields to catalogs of emission-line galaxies from with weak emission lines, but do find 69 CLU-Hα galax- previous blind emission-line surveys. ies with r-band magnitudes fainter than 18 mag with large We note that the Hα galaxy catalogs and analyses pre- Hα EWs (a median value of 45 A)˚ illustrating that a our sented here utilize single exposure images; not the final selection methods can detect fainter galaxies with strong stacked images. The stacked CLU-Hα images will produce

c 0000 RAS, MNRAS 000, 000–000 12 D. Cook et al.

Figure 11. The flux properties all galaxies with a known redshift Figure 12. The Hα flux histogram and completeness for the from NED, SDSS, or our spectroscopic follow up in our volume CLU-Hα galaxies compared to all galaxies with a known red- (z≤0.047) with an Hα EW≥7.5 A.˚ The blue symbols and his- shift from NED, SDSS, or our spectroscopic follow up in the tograms represent a subset with Σ ≥ 2.5 (candidates) and the red preliminary fields. In the top panel the unfilled, red-filled, and symbols and histograms represent a subset with Σ < 2.5 (non- blue-filled histograms represents all galaxies in the volume, all candidates). The top panel shows that the galaxies with a large galaxies in the volume with Σ < 2.5 (non-candidates), CLU- enough EW to be selected in our survey but not selected as can- Hα galaxies (Σ >= 2.5). The bottom panel shows the complete- didates tend to be the fainter objects. The bottom panel shows ness percentage compared to all known galaxies in each bin of that these galaxies not selected as candidates also tend to have Hα flux versus Hα flux. The CLU-Hα galaxies are 90% complete increased photometric scatter which require a stronger emission at ∼ 1 × 10−14 erg s−1 cm−2 and ∼ ×10−14 erg s−1 cm−2 for line to be selected by our signal-to-noise selection criteria. Σ > 2.5 and Σ > 5, respectively. a galaxy catalog with fainter detection limits and a higher boxes highlight which filter the emission line has been con- completeness than the preliminary fields presented here. The firmed. The galaxies in this mosaic span a range of Hα EW, stacking methods and analysis of the CLU-Hα images are where EW increase towards the bottom of the figure from the subject of a future study and beyond the scope of this EW=10 A˚ at the top to EW=100 A˚ at the bottom. The paper. galaxies also show a range of morphologies from compact to irregular to those showing spiral structure.

4.1 Example New Galaxies 4.2 Observable Properties of CLU-Hα Galaxies Here we present examples of CLU-Hα galaxies with no previous distance information that are now well constrained to Here we explore the observable properties of our confirmed be in the local Universe. Figure 14 presents a representative galaxy candidates in comparison to cataloged galaxies in our sample mosaic of these galaxies where the Hα emission line preliminary fields with spectroscopic information from SDSS has been spectroscopically confirmed in the filter identified or our CLU-Hα followup. Figure 15 presents the SDSS g − r by our narrow-band colors. The left image cutout in Fig- color, apparent r-band magnitude, the absolute r-band mag- ure 14 is the SDSS gri color composite, the four panels to nitude, and the Hα luminosity of all galaxies with spectro- the right are the cutouts for all four Hα filters, and the green scopic information, a subset of these galaxies with Hα EW

c 0000 RAS, MNRAS 000, 000–000 CLU I: Galaxy Catalogs from Preliminary Fields 13

Figure 13. The distribution of absolute Mr-band magnitude and redshift for all galaxies with a known redshift from NED, SDSS, or our spectroscopic follow up and CLU-Hα galaxies. The top panel presents the full redshift range of the CLU-Hα candidates while the bottom panel is a zoom-in of the local Universe (z<0.05). The vertical dotted line represents the maximum redshift limit of our survey (i.e., z=0.047), the dashed- and solid-curved lines represent the relationship between an apparent magnitude limit of 17.8 (i.e., the limit of the SDSS spectroscopic galaxy survey) and 19 mag, respectively, with redshift. The small black, larger red, larger blue dots, and X’s represent all known galaxies, Σ = 2.5 CLU-Hα galaxies, Σ = 5 CLU-Hα galaxies, and CLU-Hα galaxies with no previous distance information, respectively.

greater than 7.5 A,˚ and our CLU-Hα Σ = 2.5 galaxies. We ies in our preliminary fields have g − r colors that peak do not show the observable properties for Σ = 5 catalog around 0.75 mag largely due to the presence of early-type since their distributions are similar to the Σ = 2.5 catalog. galaxies in the Coma field. However, both the sub-sample The sub-sample of galaxies with an Hα EW greater than with Hα EW> 7.5A˚ and the CLU-Hα galaxies show a bluer 7.5 A˚ represent those above our minimum Hα EW selec- peak around 0.45 mag. The different color distributions show tion threshold for the Σ = 2.5 galaxy list. Thus, these com- that our selection methods tend to select blue, star-forming parisons will reveal the properties of the galaxies with EW galaxies; as expected. values above our limit that were not selected. Panel ‘b)’ of Figure 15 shows the apparent r-band Panel ‘a)’ of Figure 15 shows that the majority of galax- magnitude where we find median values of 16.5, 16.9, and

c 0000 RAS, MNRAS 000, 000–000 14 D. Cook et al.

Figure 14. A mosaic of a representative sample of newly discovered galaxies in the local Universe, where (from left-to-right) the image cutouts are the SDSS gri composite and followed by the four Hα filters (Hα1-Hα4). The Hα magnitudes and errors are listed above each Hα cutout and the green boxes indicate which filter the galaxy was identified (i.e., the brightest). In addition, the galaxies are sorted by Hα EW, where the lowest EW galaxies are at the top (EW∼10 A)˚ and the highest are at the bottom (EW∼100 A).˚

16.6 mag for galaxies in our preliminary fields with spec- methods have fainter apparent magnitudes with a median troscopy, the sub-sample with an Hα EW>7.5A,˚ and CLU- value at 17.2 mag. These fainter galaxies are not selected Hα galaxies, respectively. In addition, the distribution of since the signal-to-noise criteria in our method is an increas- the Hα EW>7.5 A˚ sub-sample is peaked at fainter r-band ing function of “On-Off” color towards fainter magnitudes. magnitudes between 17–18 mag when compared to the CLU- In other words, the random “On-Off” color scatter of sources Hα galaxies. This suggests that a larger fraction of galaxies due to the noise in the images is greater at fainter magni- with moderate Hα EWs that are missed by our selection

c 0000 RAS, MNRAS 000, 000–000 CLU I: Galaxy Catalogs from Preliminary Fields 15 tudes and thus require a larger Hα EW to be selected (See dex (Meidt et al. 2014; McGaugh & Schombert 2015). A con- the signal-to-noise curves in Figures 5–8). stant mass-to-light ratio of the same value for WISE 3.4µm Panel ‘c)’ of Figure 15 shows the absolute Mr mag- has been shown to yield similar stellar masses as Spitzer nitude where we find median values of –18.7, –18.5, and 3.6µm (Jarrett et al. 2013; Norris et al. 2014). We adopt the 3.4µm –18.9 mag for galaxies in our preliminary fields with spec- constant mass-to-light ratio of Υ? = 0.5 M /L ,3.4µm, troscopy, the sub-sample with an Hα EW>7.5A,˚ and CLU- where M /L ,3.4µm is the mass-to-light ratio in units of Hα galaxies, respectively. We find that the subset of galax- solar masses per the solar luminosity in the WISE 3.4µm ies with an Hα EW>7.5 A˚ and not selected in our survey filter bandpass (m ,3.4µm=3.24 mag; L ,3.4µm = 1.58 × span a range of magnitudes including both intrinsically faint 1032 erg s−1; Jarrett et al. 2013). and bright galaxies. These galaxies tend to have fainter ap- Star formation rates (SFRs) for our galaxies can be es- parent magnitudes due to their distances, where 75% are timated from many different luminosity tracers (e.g. Hα, at distances greater than 100 Mpc with a median r-band FUV, etc.), where measured luminosities are transformed magnitude of 17.2 mag. Thus, these missed galaxies with into SFRs via scaling prescriptions (e.g., Kennicutt 1998; moderate Hα EWs would require stronger Hα emission to Murphy et al. 2011). We utilize the updated scaling relation- be selected in our survey. ships of Murphy et al. (2011) which assumes a Kroupa IMF Panel ‘d)’ of Figure 15 shows the Hα luminosity where (Kroupa 2001). The Hα SFRs are derived via a combination −1 2 we find median Log LHα (erg s cm ) values of 39.1, of Hα and 22µm luminosities which account for internal dust 39.7, and 39.9 for galaxies in our preliminary fields with extinction (Calzetti et al. 2010; Murphy et al. 2011): spectroscopy, the sub-sample with an Hα EW>7.5A,˚ and CLU-Hα galaxies, respectively. This panel clearly illustrates SFR (M yr−1) = C × νL + 0.031 × νL , (9) that a large fraction (>50%) of galaxies with spectroscopy Hα,corr Hα 22µm in our preliminary fields have weak Hα emission (i.e., small where νL are the observed monochromatic luminosities of Hα EWs). We note that the early-type galaxies with Hα line both Hα and the WISE4 band at 22µm in ergs per second non-detections are not plotted in this panel. A comparison and C=5.37×10−42 (Murphy et al. 2011). between the subset of galaxies with EW>7.5 and our CLU- The two panels of Figure 16 show the Hα-derived SFR Hα galaxies shows that the CLU-Hα galaxies have a larger and stellar mass distributions for the same samples as in median value by 0.2 dex. Thus, our methods tend to select Figure 15. Panel ‘a)’ shows the SFR histograms where we −1 galaxies with the intrinsically higher Hα luminosities. We find median Log SFR (M yr ) values of –1.0, –0.7, and note that the moderate EW galaxies missed by our selec- –0.5 for the galaxies in our preliminary fields with spec- tion methods tend to be fainter with a median r-band mag- troscopy, the sub-sample with an Hα EW>7.5A,˚ and CLU- nitude of 17.1 mag, and would require larger Hα emission Hα galaxies, respectively. Our selection methods tend to se- to be selected. lect galaxies with higher SFRs. However, we are able to recover some galaxies with moderately low SFRs near Log −1 SFR (M yr )∼−2. 4.3 Physical Galaxy Properties Panel ‘b)’ of Figure 16 shows the stellar mass his- Here we present the physical properties of the galaxy sample tograms where we find median Log M? (M ) values of 9.5, by cross-matching against the ALLWISE 3.4µm and 22µm 9.4, and 9.6 for the galaxies in our preliminary fields with ˚ fluxes. These data will provide stellar masses (M?) and ex- spectroscopy, the sub-sample with an Hα EW>7.5A, and tinction corrections due to dust. In addition, we use the CLU-Hα galaxies, respectively. The median values suggest spectroscopic Hα fluxes of our survey to derive SFRs. The that our methods select galaxies of roughly the same masses Hα fluxes have been corrected for Milky Way extinction via as other galaxies in the preliminary fields. We also find that the prescription of Schlafly & Finkbeiner (2011). the CLU-Hα galaxies span a range of stellar mass (8 . Log The stellar masses are derived from mass-to-light ratios M?(M ) to . 11) suggesting that CLU-Hα will recover some (Υ?) using the WISE 3.4µm fluxes. We utilize the fluxes lower-mass dwarfs as well as larger spirals. However, CLU- derived from ALLWISE catalog profile fitting photometry, Hα will miss a significant fraction of dwarfs at greater dis- thus these fluxes should encompass each galaxies’ full ra- tances. dial extent. The WISE 3.4µm bandpasses provides a robust tracer of a galaxy’s stellar mass as this light is dominated 4.4 Comparison to Previous Blind Surveys for by an older stellar population (which make up the majority Active Galaxies of a galaxy’s stellar mass) and is less affected by attenuation from dust than shorter wavelengths. There have been several blind emission-line and UV-excess Many studies over the past few years have made com- surveys undertaken prior to CLU-Hα including the Markar- parisons between a variety of observationally derived stellar ian Survey (Markarian, Lipovetskii & Stepanian 1981 and masses (e.g., baryonic Tully-Fisher relationship and resolved references therein), the Case Northern Sky Survey (Pesch star color-magnitude diagrams) and luminosities in Spitzer & Sanduleak 1983), the KPNO International Spectroscopic 3.6µm and WISE 3.4µm bandpasses (Oh et al. 2008; Es- Survey (KISS; Salzer et al. 2000), the University of Michin- kew, Zaritsky & Meidt 2012; Barnes et al. 2014; McGaugh gan survey (UM; MacAlpine, Smith & Lewis 1977a,b; & Schombert 2014; Meidt et al. 2014; Norris et al. 2014; MacAlpine, Lewis & Smith 1977; MacAlpine & Lewis 1978; McGaugh & Schombert 2015; Querejeta et al. 2015). The MacAlpine & Williams 1981), and the Hamburg/SAO Sur- results of these comparisons show that a constant Spitzer vey for Emission-Line Galaxies (HSS; Ugryumov et al. 1999). 3.6µm Υ? of 0.4 − 0.55 M /L provides a robust estimation Here we compare our galaxy catalogs (Σ > 2.5 and Σ > 5) of a galaxies stellar mass with a relatively low error of ∼0.1 to the objects in the first three surveys since there is no

c 0000 RAS, MNRAS 000, 000–000 16 D. Cook et al.

Figure 15. The observable property histograms of our Σ = 2.5 galaxies in comparison to galaxies with Hα EW information from SDSS or our CLU-Hα followup. The unfilled, red-filled, and blue-filled histograms represent all galaxies with existing spectroscopy in our preliminary fields, a subset of these galaxies with an EW>7.5 A,˚ and our Σ = 2.5 galaxies, respectively. Panel a) shows the g − r optical colors, panel b) shows the apparent r-band magnitudes, panel c) shows the absolute r-band magnitudes, and panel d) shows the Hα derived luminosities. We find that the galaxies not selected by our methods with an EW greater than our selection limit have fainter apparent magnitudes near 18 mag. These fainter galaxies are not selected since our signal-to-noise criteria is an increasing function of “On-Off” color towards fainter magnitudes.

overlap between our preliminary fields and both UM and Finally, the KISS survey used an objective prism and line HSS. selection (Hα) but implemented modern CCD detectors to The Markarian (i.e., the first Byurakan; Markarian & survey fainter sources (20-21 mag) and found 2425 galaxies Stepanian 1983; Markarian et al. 1989; Petrosian et al. 2007) in ∼ 200 square degrees. The use of CCDs and line selec- survey used an objective prism and photographic plates to tion methods facilitated deeper surveys and the detection of find a total of 1544 UV-excess galaxies via a visual search galaxies with a wider range of properties since the Hα line for steep UV continuum slopes in their low-dispersion spec- (compared to UV slope and [OIII] line selection) will be tra. This was the first systematic search for active galax- stronger in more massive galaxies. ies and covered 17,000 square degrees down to a continuum The CLU-Hα survey utilizes CCD detectors and brightness of 17.5 mag in the V-band. The Case survey used narrow-band imaging for low resolution line selection (Hα) a combination of steep UV continuum slopes and emission to increase the survey area (3π for the full survey) and to en- line selection ([OIII]) to find more galaxies per area resulting sure the detection of galaxies with a wide range of properties. in 2339 galaxies in ∼ 1200 square degrees down to 18 mag. Using single image exposures in preliminary fields, we found

c 0000 RAS, MNRAS 000, 000–000 CLU I: Galaxy Catalogs from Preliminary Fields 17

Figure 16. The physical property histograms of of our Σ = 2.5 galaxies in comparison to galaxies with Hα EW information from SDSS or our CLU-Hα followup. The unfilled, red-filled, and blue-filled histograms represent all galaxies with existing spectroscopy in our preliminary fields, a subset of these galaxies with an EW>7.5 A,˚ and our Σ = 2.5 galaxies, respectively. We find that our selection methods tend select galaxies with higher SFRs, and can miss low-SFR or low-mass galaxies depending on their distance, and consequently their apparent magnitude.

258 confirmed galaxies with Σ > 2.5 in 100 deg2 and can CLU-Hα found an additional 247 objects in the overlapping detect the continuum of sources down to 18.5 mag. Thus, volume. The one Markarian galaxy not selected in CLU- the CLU survey depth and galaxy density are in between Hα has a star within 100 of the galaxy center and was misla- previous surveys that used photographic plates (Markarian beled as a star in our catalog. We note that this galaxy still and Case) and modern CCDs (KISS). exhibits a Σ value of 15. Figure 17 shows the sky regions covered by the CLU- A comparison with the Case galaxies shows that we suc- Hα preliminary fields presented in this paper and each of cessfully recover 14 of the 18 galaxies (78%). However, CLU- the three comparison surveys with spatial overlap: Markar- Hα found an additional 30 objects in the overlapping survey ian (Petrosian et al. 2007), Case (Pesch & Sanduleak 1983; regions. Two of the Case galaxies missed by CLU-Hα have Sanduleak & Pesch 1984; Pesch & Sanduleak 1986; Sand- small EW values below 2 A.˚ The third galaxy missed by uleak & Pesch 1987; Pesch & Sanduleak 1988, 1989; Sand- CLU-Hα is an extended galaxy whose central region shows uleak & Pesch 1989, 1990; Pesch, Sanduleak & Stephenson a small Σ value; however, we did find several of this galaxy’s 1991), and KISS (Salzer et al. 2000, 2001; Gronwall et al. H ii regions with high Σ values. The last galaxy missed by 2004; Jangren et al. 2005). In addition, the bottom panel of CLU-Hα shows two nearby H ii regions where our photo- Figure 17 is a zoomed-in version covering our preliminary metric aperture is located between two clumps and thus is fields in more detail between an RA of 180–270 degrees. We missing some of the Hα flux. first spatially crossmatched CLU-Hα to the galaxies in each A comparison with the KISS galaxies shows that we suc- of the catalogs that are spatially coincident with our prelim- cessfully recover 42 of the 80 galaxies (53%). However, CLU- inary fields (within 400), and have limited these catalogs to Hα found an additional 24 objects in the overlapping survey galaxies with a redshift below 0.047 (i.e., redshift probed by regions. Nearly a third of the KISS galaxies missed by CLU- our survey). We also note that one of our CCD chips (#3) Hα (N=12) were below our detection limits of 18.5 mag. has been nonfunctional for all of our pointings since the be- The remaining KISS galaxies were missed by CLU for the ginning of the survey. In addition, the two lower-left chips following reasons: 8 had low Hα EWs less than 7.5 A,˚ 6 for the field ’p3967’ (center near RA=195, Dec=28) in the were labeled as stars, and 12 had moderate Hα EWs com- Hα4 filter had poor data quality and could not be reduced. bined with fainter magnitudes between 17.5–18.5 mag. The We remove galaxies that overlap with chip 3 in all pointings 12 missed galaxies with moderate EWs were cut in our sur- and the poor data chips for ’p3967’ in this comparison. The vey since their fainter magnitudes required higher EWs to number of galaxies from the three comparison catalogs with exhibit a significant narrow-band color. a redshift less than 0.047 and inside our preliminary fields The CLU-Hα survey is able to recover the majority of is 8, 18, and 80 for the Markarian, Case, and KISS surveys, the galaxies in both the Markarian and Case surveys, and respectively. was able to find additional galaxies. Our comparison with A comparison with the Markarian galaxies shows that the KISS survey shows that CLU-Hα recovers roughly half we successfully recover 7 of the 8 galaxies (88%). However, as many emission-line galaxies as KISS due to our brighter

c 0000 RAS, MNRAS 000, 000–000 18 D. Cook et al. limits. However, CLU-Hα will cover a much larger area of Hα, MASH; Parker et al. 2006) found a few hundred newly the sky (3π sr) compared to all previous emission-line galaxy discovered planetary nebulae at galactic latitudes of 5 < surveys to a moderate depth. Thus, the CLU-Hα survey will |b| < 15 degrees (≈ 4000 deg2) with a detection limit of discover many emission-line galaxies across the sky. 5 × 10−16erg s−1cm−2. Despite the higher selection limit of our survey (1 × 10−14erg s−1cm−2), we anticipate finding 10s-to-100 new PNe in the area not covered by IPHAS 2 5 DISCUSSION and MASH (≈ 3500 deg ) at intermediate galactic latitudes assuming the same distribution of PNe as in the southern In this section we examine interesting candidates found hemisphere. in the preliminary fields of the Hα survey where we find In the CLU-Hα sample, there are 9 BCDs where 7 have new BCDs, a newly discovered green pea, a new Seyfert distances measured for the first time in this survey. Panel b) 1 galaxy, and a known planetary nebula. We also put the of Figure 18 shows the image cutouts of an example BCD CLU-Hα galaxy properties into the context of previously with new distance measurements, where the source is bright- established relationships between different galaxy physical est in the fourth Hα filter. The second panel of Figure 19 properties (e.g., the star-forming “Main Sequence”). The shows the spectrum of the BCD where the redshifted wave- majority of the CLU-Hα galaxies show physical properties length of the strong Hα line confirms the identification in similar to normal star-forming galaxies; however, several ex- the fourth filter. We note that the Hα emission line has a treme galaxies (i.e., BCDs) show deviations from previously wavelength that is only a few angstroms greater than the established galaxy trends. We end this section with a discus- separation between the third and fourth filter, and is clearly sion of how the CLU-Hα survey can help focus the search brighter in the fourth filter image. The spectrum of the ex- for the electromagnetic counterparts to gravitational wave ample BCD is representative of the other BCDs, where both events. the [OIII]λ4959, λ5007 and Hα lines exhibit strong emission lines: EWs of 418 A˚ and 446 A˚ for [OIII]λ4959, λ5007 and Hα, respectively. We measure the 12+Log(O/H) metallicity 5.1 Interesting Candidates of 8.02 from O3N2 methods (Pettini & Pagel 2004) and a −1 Amongst our emission-line candidates we find interesting ex- dust corrected SFR of 0.41 M yr . treme objects some of which have no previous distance infor- The SDSS and CLU-Hα images of the green pea are mation. In figure 18 we present a mosaic of four interesting shown in panel c) of Figure 18. The spectrum of this ob- candidates, where panel a) shows the known planetary neb- ject is shown in the third panel of Figure 19 where the ula, panel b) shows one of the BCDs, panel c) shows the [OIII]λ4959, λ5007 lines confirm the correct identification in green pea, and panel d) shows Seyfert 1 galaxy. Due to the the Hα3 filter. Both the [OIII]λ4959, λ5007 and Hα lines large “On-Off” magnitudes of the planetary nebula, the cen- exhibit strong emission lines: EWs of 510 A˚ and 560 A˚ for tral pixels in the top panel of Figure 18 in the first Hα filter [OIII]λ4959, λ5007 and Hα, respectively. We have measured were masked to provide a more consistent background and metallicity of 12+log(O/H) = 8.09 via the O3N2 method. −15 −1 −2 visual comparison across the four Hα images. The Hα flux of 9. × 10 erg s cm which corresponds −1 The planetary nebula is PN H 4-1 and is located at a to a dust corrected SFR of 24 M yr given the measured high Galactic lattitude: b = 88.14757◦, l = 49.3065◦ (Haro redshift of 0.337 and H0 of 72 km/s/Mpc. 1951). Panel a) of Figure 18 shows the SDSS gri color and The image cutouts of the new Seyfert 1 galaxy found our 4 Hα filters (from left-to-right, respectively). PN H 4- in our survey of preliminary fields is shown in panel d) 1 shows a large flux excess in Hα1 compared to Hα2 (i.e., of Figure 18 where the object is brightest in the third “On-Off”) equal to three magnitudes (∆mag = −3.05). The Hα filter. In addition, the spectrum of this object is shown measured spectroscopic Hα line flux and EW are 1.21 × in the bottom panel of Figure 19 which shows strong 10−12erg s−1cm−2 and 1200 A,˚ respectively. The spectrum [OIII]λ4959, λ5007 lines confirming the correct identifica- of this PN is shown in top panel of Figure 19, and exhibits tion in our third filter. Furthermore, the broadened Hα and emission lines similar to other known planetary nebulae. Hβ emission lines clearly indicate the presence of a strong The detection of this planetary nebula is interesting AGN which allows us to classify this object as a Seyfert 1 since our choice of preliminary fields avoided the Galac- galaxy. tic plane and PN H 4-1 is located at a high Galactic latitude (b ∼88◦). We anticipate that the CLU-Hα survey 5.2 Galaxy Trends can be used as a discovery engine for new planetary nebulae at intermediate galactic latitudes. This expectation is In this section we show where the CLU-Hα galaxies are lo- based on the galactic latitude limits of previous emission- cated on established trends of star-forming galaxies. Pre- line galaxy and Galactic Plane surveys in the northern sky. vious studies of star-forming galaxies have found a rela- Previous galaxy surveys (Markarian, Case, UM, KISS, HSS, tively tight correlation between the current SFR and the to- etc.) avoided the galactic plane above |b| > 20◦, while the tal stellar mass for both local Universe and higher redshift largest-area galactic plane survey in the northern hemi- galaxy samples: the galaxy “Main Sequence.” This trend sphere was limited to |b| < 5◦ (The INT Photometric can provide insights into how galaxies evolve over time since Hα Survey, IPHAS; Drew et al. 2005). Thus, there is a the normalization increases with redshift (e.g., Heinis et al. gap at intermediate galactic latitudes that has not been 2014). We use this relationship to illustrate the distribution uniformly searched for emission-line sources. In addition, of CLU-Hα galaxy properties. the large-area galactic plane search for planetary nebulae in Figure 20 shows the dust-corrected Hα-SFR versus stel- the southern hemisphere (The Macquarie/AAO/Strasbourg lar mass (M?) for the CLU-Hα galaxies, where the symbols

c 0000 RAS, MNRAS 000, 000–000 CLU I: Galaxy Catalogs from Preliminary Fields 19

Figure 17. Top: The locations of catalogued galaxies below z=0.047 in the CLU-Hαpreliminary ﬁelds as well as the KISS, Case, and Markarian surveys. Bottom: Here we highlight a smaller region of the spring sky to show more clearly where our preliminary ﬁelds overlap with the KISS, Case, and Markarian surveys.

are the same as those in Figure 13. The blue and green 2014b; Schawinski et al. 2014). Figure 21 shows the sSFR lines represents the relationships found in the LVL sample versus the SDSS g − r color of the CLU-Hα galaxies, where (D<11 Mpc; Cook et al. 2014b) and SDSS sample (z∼ 0.1; the symbols are the same as that in Figure 20. We find Peng et al. 2010). The majority of the CLU-Hα galaxies fol- that the CLU-Hα galaxies follow a similar trend to the LVL low the previous “Main Sequence” relationship and span a galaxies (Cook et al. 2014b), but that several CLU-Hα BCDs wide range in both SFR and stellar mass from low-mass, tend to have sSFRs (Log(sSFR) as high as −8.7) and bluest low-SFR dwarfs to high-mass, high-SFR spirals. optical colors due to the large volume sampled. The CLU- The BCDs found in the CLU-Hα sample show high Hα green pea galaxy shows a sSFR and color similar to that SFRs given their stellar masses illutstrating that these of the other CLU-Hα galaxies. A future CLU-Hα study will galaxies are undergoing an episode of enhanced star forma- include larger numbers of BCDs and green pea galaxies to tion. The new green pea, on the other hand, shows good study this relationship. agreement with the SDSS “Main Sequence”relationship suggesting that this galaxy is not undergoing an enhanced pe- riod of star formation; this is contradictory to previous stud- 5.3 Electromagnetic Counterparts to ies which find that green peas show signs of significant star Gravitational Wave Events formation activity (Cardamone et al. 2009; Izotov, Guseva The detection of gravitational waves via the Laser Inter- & Thuan 2011). However we do not have enough statistics in ferometer Gravitational Wave Observatory (LIGO; Abbott this study to draw meaningful conclusions as to the connec- et al. 2016) allows for a new way of observing the Universe tion between BCDs and green peas. Such an analysis is left and thus provides new tools with which to understand fun- for a future CLU-Hα study that will collect a larger sample damental physics. Associating gravitational waves and elec- of both types of objects. tromagnetic counterparts will have dramatic impacts on our We also explore a relationship between physical and ob- understanding of the Universe. However, the size of the 90% served properties of galaxies, where local Universe studies sky localizations of LIGO GW events have ranged from 30- have found that the specific SFR (sSFR≡SFR/M?) forms a 1000 square degrees on the sky making the search for EM relatively tight trend with optical colors (e.g., Cook et al. counterparts a challenge.

c 0000 RAS, MNRAS 000, 000–000 20 D. Cook et al.

Figure 18. A mosaic of some interesting candidates found in our preliminary ﬁelds similar to Figure 14. Panels a-d shows the known planetary nebula, an example of a new BCD, the new green pea, and the new Seyfert 1 galaxy, respectively. In addition, the thin galaxy just above center in panel b) is also newly discovered CLU-Hα galaxy with an Hα EW=80 A.˚

Significant efforts of large area surveys (e.g., Kasliwal eighth of the LIGO sensitivity volume out to 200 Mpc. In ad- et al. 2016; Coulter et al. 2017; Evans et al. 2017; Hallinan dition, several new or updated spectroscopic galaxy surveys et al. 2017; Kasliwal et al. 2017; Troja et al. 2017) have have published secure distances via redshifts for new galax- scanned the most probable sections of gravitational wave ies in the local Universe (e.g., SDSS, ALFALFA, 6dFRGS; localizations in search of an electromagnetic counterpart. Alam et al. 2015; Haynes et al. 2011; Jones et al. 2009); However, the efficiency of this effort can be greatly enhanced thus, leaving previous EMGW galaxy catalogs less complete by utilizing the locations of known galaxies in the LIGO than previously estimated. There are efforts which utilize in- sensitivity volume for neutron star mergers (D≤200 Mpc; creased numbers of galaxies with photometric redshifts (e.g., Aasi et al. 2015). Targeted follow-up observations of likely GLADE5); however, we focus on galaxy catalogs with secure host galaxies can narrow down the search area by a factor distance measurements. The lack of an EMGW catalog that of 100 (Nissanke, Kasliwal & Georgieva 2013; Gehrels et al. extends to the full LIGO sensitivity volume and the exis- 2016). tence of new galaxies found in the local Universe motivate the construction of a new compiled galaxy catalog to be used Previous efforts have been made to provide lists of in the search for EM counterparts to GW events. galaxies specifically designed for EMGW follow-up (Kop- In an effort to provide the most complete list of galax- parapu et al. 2008; White et al. 2011), where the latest catalog from White et al. (2011) shows a B-band luminosity completeness of 65% at 100 Mpc. However, the galaxies in both previous catalogs were limited to 100 Mpc which is an 5 http://aquarius.elte.hu/glade/

c 0000 RAS, MNRAS 000, 000–000 CLU I: Galaxy Catalogs from Preliminary Fields 21

3.5 10-13 × planetary nebula [Oiii] 3 10-13 × Hα 2.5 10-13 × 2 10-13 × [Nii] 1.5 10-13 Hα × [Nii] 1 10-13 × 6600 6800 Hβ 0.5 10-13 × Hγ 0 4 10-15 × BCD [Oiii]

3 10-15 Hα × Hα

2 10-15 ×

[Nii] 1 10-15 Hβ × 6600 6800 Hγ [Sii] 0

green pea Hα 1.5 10-15 × )

1 [Oiii] − s 2

− 1 10-15 × [Oiii] [Oiii]

(erg cm -15 Hβ

λ 0.5 10

F × [Oii] [Sii] Hβ 6600 6800

0 7 10-16 × Seyfert 1 [Oiii] 6 10-16 × [Oiii] 5 10-16 × 4 10-16 Hα × [Oiii] 3 10-16 × Hβ 2 10-16 × [Oii] 6600 6800 Hβ [Sii] 1 10-16 × 0 4000 4500 5000 5500 6000 6500 7000 7500 8000 8500 9000 Wavelength (A)˚

Figure 19. The Palomar DBSP spectra for the interesting candidates found in our preliminary fields where (from top-to-bottom) the panels display the known planetary nebula, an example BCD, the green pea, and the Seyfert 1 galaxy. Strong nebular lines are labeled in each panel. The inserts in each panel show a zoomed-in window where the vertical dashed lines represent the wavelength range covered by our narrow-band filter pairs (6500−6692 A˚ and 6692−6900 A).˚ The dotted lines represent the separation between the filters within the filter pairs.

ies with measured distances in the LIGO sensitivity volume, 2014), Extragalactic Distance Database8 (EDD; Tully et al. our team has compiled a catalog of all known galaxies out to 2009), the Sloan digital sky survey DR12 (SDSS; Alam et al. 200 Mpc, hereafter referred to as CLU-compiled. The galax- 2015), The 2dF Galaxy Redshift Survey (6dFRGS; Jones ies were taken from existing galaxy databases: NASA/IPAC et al. 2009), and The Arecibo Legacy Fast ALFA (ALFALFA Extragalactic Database (NED)6, Hyperleda7 (Makarov et al. Haynes et al. 2011). Distances based on Tully-Fischer methods were favored over kinematic (i.e., redshift) distances;

6 https://ned.ipac.caltech.edu 7 http://leda.univ-lyon1.fr 8 http://edd.ifa.hawaii.edu

c 0000 RAS, MNRAS 000, 000–000 22 D. Cook et al.

however, the majority of the distances are based upon redshift information. The catalog contains ∼234,500 galaxies with existing distances less than 200 Mpc. In addition to distances, the catalog also contains compiled photometric information. We have cross-matched (within a 400 separation) the CLU-compiled catalog with GALEX all sky (Martin et al. 2005), WISE all sky (Wright et al. 2010), and SDSS DR12 (Alam et al. 2015) surveys to obtain fluxes from the ultraviolet (UV) to the infrared (IR). We find ∼154,200 matches for GALEX FUV, ∼216,000 for WISE 3.4 and 22µm, and ∼100,400 for SDSS r-band. With the UV and IR fluxes, we have also derived physical properties (SFRs, stellar masses, and dust extinction) for the CLU- compiled galaxies via similar methods as the CLU-Hα survey in §4.3. The CLU galaxy catalog is being utilized by the GROWTH (Global Relay of Observatories Watching Tran- 9 Figure 20. The “Main Sequence” of star-forming galaxies for sients Happen ) collaboration to search for EM counter- CLU-Hα galaxies in the preliminary fields. The red dots, blue parts to gravitational wave (GW) events across the electro- dots, large X’s represent Σ = 2.5 CLU-Hα galaxies, Σ = 5 CLU- magnetic spectrum. On August 17th 2017 GWs from two Hα galaxies, and CLU-Hα galaxies with no previous distance in- merging neutron stars were observed in both the LIGO and formation. The blue and green lines represents the relationships Virgo detectors (hereafter GW170817; Abbott et al. 2017a) found in the LVL sample (D<11 Mpc; Cook et al. 2014b) and and is the first GW event to have an electromagnetic coun- SDSS sample (z∼ 0.1; Peng et al. 2010). The green soliddot- terpart found (Abbott et al. 2017b). Shortly after the GW ted line transition represents the lower end of the stellar mass event, several collaborations around the world began a large range probed by the SDSS survey. The error bars near the legend represent the median errors of all data points. We find that the multi-wavelength observational campaign to search for the majority of the CLU-Hα galaxies show agreement with the local electromagnetic counterpart (e.g., Coulter et al. 2017; Evans Universe “Main Sequence”. In addition, the blue and green dia- et al. 2017; Hallinan et al. 2017; Kasliwal et al. 2017; Troja monds represent the BCD and green pea galaxies found in our et al. 2017). Our team crossmatched the CLU-compiled cat- preliminary fields. These extreme galaxies show higher SFRs for alog against the LIGO-Virgo volume (initially 31 square de- their given stellar mass. grees on the sky at a distance of 40±8 Mpc; Abbott et al. 2017b) and found 49 galaxies (see; Kasliwal et al. 2017). In addition, since the physical properties of these galaxies were determined in advance by our team, we were able to promptly prioritize these galaxies by stellar mass and pub- lish this list on the Gamma-ray Coordinate Network (GCN; Cook et al. 2017). The EM counterpart was later found in NGC4993 (Coulter et al. 2017), which was the third highest priority galaxy on our published list. We note that the CLU galaxy catalog produced the only published GCN to correctly identify NGC4993 in the localization volume of GW170817. Future efforts to search for the EM counterparts to GW events using CLU will include new galaxies found in the CLU-Hα survey. Next, we examine the basic properties of our CLU- compiled catalog and investigate how the new galaxies from the CLU-Hα survey will contribute to the full CLU catalog. Figure 22 shows the observable and physical property histograms of both the CLU-compiled galaxies (open histogram) and CLU-Hα galaxies in the preliminary fields

Figure 21. The specific star formation rate (sSFR≡SFR/M?) (blue-filled histogram), where all histograms have been nor- versus g − r optical color, where sSFR is an indicator for the in- malized to the peak. tensity of star formation. The error bars near the legend represent Panel a) shows the SDSS g − r color histograms, where the median errors of all data points. We find that the BCDs show the CLU-compiled sample shows a double peak while the the highest star formation rate intensities (sSFR) and tend to CLU-Hα sample shows a single peak that overlaps with the have the bluest colors. The new green pea galaxy shows a sSFR CLU-compiled blue peak. The double peak is due to the similar to other normal star-forming galaxies. dichotomy between blue star-forming galaxies and red early- type galaxies with little-to-no star formation. The absence of a red peak in the CLU-Hα sample shows that the CLU-

9 http://growth.caltech.edu

c 0000 RAS, MNRAS 000, 000–000 CLU I: Galaxy Catalogs from Preliminary Fields 23

Figure 22. A Comparison observable and physical properties between the CLU-compiled and CLU-Hα galaxy samples. Panels a, b, c, and d show the histograms of g − r color, absolute Mr-band magnitude, FUV-derived SFR, and stellar mass, respectively. We ﬁnd that CLU-Hα will likely add galaxies that tend to be blue in optical color and high in SFR to existing catalogs of galaxies; however, will miss more massive galaxies that are red in optical color (i.e., red-sequcence galaxies).

Hα survey is not sensitive to early-type galaxies. This can galaxies are already in CLU-compiled given the inclusion of also be seen in panel c) which shows the FUV-derived SFRs, SDSS galaxy spectra. where the CLU-Hα galaxies are peaked at 0.3 dex higher We will add CLU-Hα galaxies found in our narrow-band than that of CLU-compiled. Thus, the galaxies added by survey with no previous distance information to our CLU- the CLU-Hα survey will be biased towards bluer galaxies compiled catalog to produce a more complete census of the with recent star formation. local Universe. This combined catalog will then be utilized to focus the search for future electromagnetic counterparts Panel b) shows the absolute magnitude M histogram, r to gravitational wave events. where both CLU-compiled and CLU-Hα samples show two peaks: one for intrinsically bright galaxies (Mr ∼-19 mag) and one for fainter galaxies (M ∼ −14 mag). The re- r 5.4 Future Improvements duced number of low luminosity CLU-Hα galaxies suggests that the CLU-Hα survey will not probe the same number Future improvements to maximize the completeness and densities of dwarfs found in the volume out to 200 Mpc. minimize the contamination of the CLU-Hα catalog include: There is also a deficit of intrinsically bright CLU-Hα galax- (i) Stacking all Hα images to produce a deeper Hα source ies (Mr . −20 mag) compared to CLU-compiled. This can catalog and a resulting deeper galaxy catalog. These data also be seen in panel d) which shows the stellar mass his- will not only provide deeper images from which to find more tograms, where the CLU-compiled sample peaks at 0.5 dex galaxies, but will also facilitate removal of cosmic rays and higher in M?. An examination of the morphologies of the chip defects. (ii) Extended source photometry; (iii) Machine galaxies brighter than –20 mag shows that 60% have RC3 learning for star-galaxy separation; (iv) Image subtraction T-types less than 1 (or S0 and earlier morphologies). Thus, of “On” and “Off” filters. the CLU-Hα survey will not add massive early-type galaxies In addition, efforts to use machine learning for can- to existing catalogs. However, the majority of these luminous didate selection are currently underway. The preliminary

c 0000 RAS, MNRAS 000, 000–000 24 D. Cook et al.

CLU-Hα survey fields are critical to our machine learning However, CLU-Hα will cover a much larger area of the sky efforts as they establish a training set of objects with known (3π sr) compared to all previous emission-line galaxy surveys redshifts. We have run preliminary tests on how well our to a moderate depth. Thus, the CLU-Hα survey can be used machine learning algorithms perform the tasks of classifying as for the discovery of many emission-line galaxies across the objects as simply within 200 Mpc and classifying objects as sky. within one of our narrow-band filters. Preliminary results We have cross-matched the resulting CLU-Hα galax- are promising, showing that we can increase our complete- ies to GALEX and WISE to derive the physical proper- ness by using our current metrics along with additional in- ties (extinction corrected SFR and stellar mass). We find formation (e.g., Pan-STARRS, WISE, GALEX magnitudes) that the CLU-Hα galaxies span a range in both SFR and 8 as features in our machine learning algorithms (Zhang et al. stellar mass including dwarf galaxies (i.e., M? ∼ 10 M −2 −1 2018; in prep). and SFR∼ 10 M yr ) as well as larger spirals (i.e., 11 1 −1 M? ∼ 10 M and SFR∼ 10 M yr ). We find that the majority of the CLU-Hα galaxies show agreement with galaxy “Main Sequence” trends found in the local Universe. 6 SUMMARY However, we do find some extreme galaxies that lie above In this paper we have presented the Census of the Local the “Main Sequence” trend and some of the highest sSFRs Universe (CLU) emission-line (Hα) galaxy survey. The CLU- in the CLU-Hα sample. Hα survey has imaged 26,470 square degrees of the north- Within the 14 preliminary fields, we find several inter- ern sky above −20◦ declination using four narrow-band fil- esting objects with no previous distance information: 7 blue ters with a FWHM of 75−90 A˚ and a wavelength range of compact dwarfs, 1 green pea, and a Seyfert 1 galaxy; we also 6525−6878 A˚ (out to z=0.047). The observations utilize 3 identified a known planetary nebula. The existence of these spatially staggered grids where each grid has 3626 fields. objects in our preliminary fields (in just 0.3% of the full The first two filters cover the galactic plane and the last survey) exemplifies that the full CLU-Hα survey will serve ◦ two filters avoid the galactic plane (|b| & 3 ). The analysis as a discovery machine for a wide variety of objects in our of CLU-Hα fields in this study utilize only 14 preliminary own Galaxy and for extreme galaxies out to intermediate fields (≈100 deg2) in one of the spatially staggered grids; redshifts. however, future studies will examine the full ≈3π area with Finally, the CLU-Hα galaxies with newly constrained 3 stacked images on nearly every point in the CLU-Hα sur- distances will be added to existing galaxy catalogs to focus vey area. the search for the electromagnetic counterparts to gravita- In the 14 preliminary fields we implemented a widely tional wave events. We have constructed a list of galaxies used signal-to-noise selection method to quantify the pres- with distances out to 200 Mpc (CLU-compiled) and will ence of an emission-line via narrow-band color excess signifi- add the CLU-Hα galaxies to this list. Our team has already cance (Σ). In addition, we undertook a spectroscopic follow- successfully applied the CLU-compiled catalog towards this up campaign at Palomar Observatory and WIRO to obtain goal by publishing a prioritized list of galaxies within the redshifts of all galaxy candidates with Σ ≥2.5 and no previ- neutron star merger event detected by the LIGO and Virgo ous redshift information (N=334). These redshifts allow us detectors on August 17th 2017 (GW170817). Within hours to explore the composition of our galaxy candidates at dif- of the LIGO trigger, our galaxy list was published on the ferent Σ cuts. There are 290 galaxies whose Hα emission line Gamma-ray Coordinates Network (GCN) and the electro- has been confirmed in our filters with 61% contamination for magnetic counterpart to GW170817 was later found in the Σ ≥2.5, and 151 confirmed galaxies and 12% contamination third highest priority galaxy on our list. for Σ ≥5. We conclude that Σ = 2.5 cut produces a more complete catalog of galaxies in our local volume but has higher contamination while a Σ = 5 cut produces a high fi- REFERENCES delity galaxy catalog with little contamination. We release a table of all candidates above a Σ = 2.5 cut in the electronic Aasi J. et al., 2015, Classical and Quantum Gravity, 32, version of this publication. 074001 In addition, we use the 14 preliminary fields to examine Abazajian K. N. et al., 2009, ApJS, 182, 543 the limits of CLU-Hα survey. The median 5σ detection limit Abbott B. P. et al., 2016, Physical Review Letters, 116, for a point source in the narrow-band imaging is 18.5 mag. A 061102 comparison of Hα fluxes derived from narrow-band imaging Abbott B. P. et al., 2017a, Physical Review Letters, 119, and spectroscopy show good agreement for the majority of 161101 galaxies, but show some contamination from [NII] lines near Abbott B. P. et al., 2017b, Physical Review Letters, 119, the Hα lines. A comparison to galaxies with spectroscopic 161101 redshifts and fluxes in our preliminary fields shows that we Alam S. et al., 2015, ApJS, 219, 12 recover 90% of the Hα flux at 1×10−14 erg s−1 cm−2 and 4× Amor´ınR. O., Pérez-Montero E., V´ılchez J. M., 2010, ApJ, 10−14 erg s−1 cm−2 for Σ cuts above 2.5 and 5, respectively. 715, L128 The CLU-Hα survey is able to recover the majority of Barnes K. L., van Zee L., Dale D. A., Staudaher S., Bullock the galaxies in both the Markarian and Case surveys, and J. S., Calzetti D., Chandar R., Dalcanton J. J., 2014, ApJ, was able to find an additional ∼250 galaxies when compared 789, 126 to the Markarian survey. Our comparison with the KISS Bellm E. C., Sesar B., 2016, pyraf-dbsp: Reduction pipeline survey shows that CLU-Hα will recover roughly half as many for the Palomar Double Beam Spectrograph. Astrophysics emission-line galaxies as KISS due to our brighter limits. Source Code Library

c 0000 RAS, MNRAS 000, 000–000 CLU I: Galaxy Catalogs from Preliminary Fields 25

Bertin E., 2006, in Astronomical Society of the Pacific Con- Kennicutt, Jr. R. C., 1998, ARA&A, 36, 189 ference Series, Vol. 351, Astronomical Data Analysis Soft- Kennicutt, Jr. R. C., Lee J. C., Funes, JoséG. S. J., Sakai ware and Systems XV, Gabriel C., Arviset C., Ponz D., S., Akiyama S., 2008, ApJS, 178, 247 Enrique S., eds., p. 112 Kopparapu R. K., Hanna C., Kalogera V., O’Shaughnessy Bertin E., Arnouts S., 1996, A&AS, 117, 393 R., GonzálezG., Brady P. R., Fairhurst S., 2008, ApJ, Bianchi L., Conti A., Shiao B., 2014, VizieR Online Data 675, 1459 Catalog, 2335 Kroupa P., 2001, MNRAS, 322, 231 Brinchmann J., Charlot S., White S. D. M., Tremonti C., Laher R. R. et al., 2014, PASP, 126, 674 Kauffmann G., Heckman T., Brinkmann J., 2004, MN- Law N. M. et al., 2009, Publications of the Astronomical RAS, 351, 1151 Society of the Pacific, 121, 1395 Bunker A. J., Warren S. J., Hewett P. C., Clements D. L., Lee J. C. et al., 2011, ApJS, 192, 6 1995, MNRAS, 273, 513 Lee J. C. et al., 2012, PASP, 124, 782 Calzetti D. et al., 2010, ApJ, 714, 1256 Lintott C. J. et al., 2008, MNRAS, 389, 1179 Cardamone C. et al., 2009, MNRAS, 399, 1191 Ly C., Lee J. C., Dale D. A., Momcheva I., Salim S., Stau- Chambers K. C. et al., 2016, ArXiv e-prints daher S., Moore C. A., Finn R., 2010, The Astrophysical Cook D., Van Sistine A., Singer L., Kasliwal M., 2017, Journal, 726, 109 Gamma Ray Coordinates Network Circular, 21519 MacAlpine G. M., Lewis D. W., 1978, ApJS, 36, 587 Cook D. O. et al., 2014a, MNRAS, 445, 881 MacAlpine G. M., Lewis D. W., Smith S. B., 1977, ApJS, Cook D. O. et al., 2014b, MNRAS, 445, 899 35, 203 Coulter D. A. et al., 2017, ArXiv e-prints MacAlpine G. M., Smith S. B., Lewis D. W., 1977a, ApJS, Daddi E. et al., 2007, ApJ, 670, 156 34, 95 Dale D. A. et al., 2009, ApJ, 703, 517 MacAlpine G. M., Smith S. B., Lewis D. W., 1977b, ApJS, Dale D. A., Giovanelli R., Haynes M. P., Campusano L. E., 35, 197 Hardy E., 1999, AJ, 118, 1489 MacAlpine G. M., Williams G. A., 1981, ApJS, 45, 113 Drew J. E. et al., 2005, MNRAS, 362, 753 Mahajan S., Haines C. P., Raychaudhury S., 2010, MN- Elbaz D. et al., 2007, A&A, 468, 33 RAS, 404, 1745 Eskew M., Zaritsky D., Meidt S., 2012, AJ, 143, 139 Makarov D., Prugniel P., Terekhova N., Courtois H., Evans P. A. et al., 2017, ArXiv e-prints Vauglin I., 2014, A&A, 570, A13 Finkelstein S. L. et al., 2012, ApJ, 758, 93 Markarian B. E., Lipovetskii V. A., Stepanian D. A., 1981, Fujita S. S. et al., 2003, The Astrophysical Journal Letters, Astrofizika, 17, 619 586, L115 Markarian B. E., Lipovetsky V. A., Stepanian J. A., Eras- Fukugita M. et al., 2007, AJ, 134, 579 tova L. K., Shapovalova A. I., 1989, Soobshcheniya Spet- Geach J. E., Smail I., Best P. N., Kurk J., Casali M., Ivison sial’noj Astrofizicheskoj Observatorii, 62, 5 R. J., Coppin K., 2008, MNRAS, 388, 1473 Markarian B. E., Stepanian D. A., 1983, Astrofizika, 19, Gehrels N., Cannizzo J. K., Kanner J., Kasliwal M. M., 639 Nissanke S., Singer L. P., 2016, ApJ, 820, 136 Martin D. C. et al., 2005, ApJ, 619, L1 Gronwall C., Salzer J. J., Sarajedini V. L., Jangren A., McGaugh S. S., Schombert J. M., 2014, AJ, 148, 77 Chomiuk L., Moody J. W., Frattare L. M., Boroson T. A., McGaugh S. S., Schombert J. M., 2015, ApJ, 802, 18 2004, AJ, 127, 1943 Meidt S. E. et al., 2014, ApJ, 788, 144 Hallinan G. et al., 2017, ArXiv e-prints Mickaelian A. M., 2014, Multiwavelength AGN Surveys Haro G., 1951, Publications of the Astronomical Society of and Studies, 304, 1 the Pacific, 63, 144 Monet D. G. et al., 2003, AJ, 125, 984 Haynes M. P. et al., 2011, AJ, 142, 170 Murphy E. J. et al., 2011, ApJ, 737, 67 Heinis S. et al., 2014, MNRAS, 437, 1268 Nissanke S., Kasliwal M., Georgieva A., 2013, ApJ, 767, Hirschauer A. S. et al., 2016, ApJ, 822, 108 124 Izotov Y. I., Guseva N. G., Thuan T. X., 2011, ApJ, 728, Noeske K. G. et al., 2007, ApJ, 660, L43 161 Norris M. A., Meidt S., Van de Ven G., Schinnerer E., Izotov Y. I., Papaderos P., Guseva N. G., Fricke K. J., Groves B., Querejeta M., 2014, ApJ, 797, 55 Thuan T. X., 2006, A&A, 454, 137 Ofek E. O. et al., 2012, Publications of the Astronomical Izotov Y. I., Thuan T. X., 2007, ApJ, 665, 1115 Society of the Pacific, 124, 854 Izotov Y. I., Thuan T. X., Guseva N. G., 2005, ApJ, 632, Oh S.-H., de Blok W. J. G., Walter F., Brinks E., Kenni- 210 cutt, Jr. R. C., 2008, AJ, 136, 2761 Izotov Y. I., Thuan T. X., Guseva N. G., Liss S. E., 2018, Oke J. B., Gunn J. E., 1982, PASP, 94, 586 MNRAS, 473, 1956 Parker Q. A. et al., 2006, Monthly Notices of the Royal Jangren A., Salzer J. J., Sarajedini V. L., Gronwall C., Astronomical Society, 373, 79 Werk J. K., Chomiuk L., Moody J. W., Boroson T. A., Pascual S., Gallego J., Arag n Salamanca A., Zamorano J., 2005, AJ, 130, 2571 2001, Astronomy and Astrophysics, 379, 798 Jarrett T. H. et al., 2013, AJ, 145, 6 Peng Y.-j. et al., 2010, ApJ, 721, 193 Jones D. H. et al., 2009, MNRAS, 399, 683 Pesch P., Sanduleak N., 1983, ApJS, 51, 171 Kasliwal M. M. et al., 2016, ArXiv e-prints Pesch P., Sanduleak N., 1986, ApJS, 60, 543 Kasliwal M. M. et al., 2017, ArXiv e-prints Pesch P., Sanduleak N., 1988, ApJS, 66, 297 Kennicutt R. C., Evans N. J., 2012, ARA&A, 50, 531 Pesch P., Sanduleak N., 1989, ApJS, 70, 163

c 0000 RAS, MNRAS 000, 000–000 26 D. Cook et al.

Pesch P., Sanduleak N., Stephenson C. B., 1991, ApJS, 76, Hawaii, the Pan-STARRS Project Office, the Max-Planck 1043 Society and its participating institutes, the Max Planck In- Petrosian A., McLean B., Allen R. J., MacKenty J. W., stitute for Astronomy, Heidelberg and the Max Planck In- 2007, ApJS, 170, 33 stitute for Extraterrestrial Physics, Garching, The Johns Pettini M., Pagel B. E. J., 2004, MNRAS, 348, L59 Hopkins University, Durham University, the University of Querejeta M. et al., 2015, ApJS, 219, 5 Edinburgh, the Queen’s University Belfast, the Harvard- Rau A. et al., 2009, PASP, 121, 1334 Smithsonian Center for Astrophysics, the Las Cumbres Ob- Salim S. et al., 2007, ApJS, 173, 267 servatory Global Telescope Network Incorporated, the Na- Salzer J. J. et al., 2000, AJ, 120, 80 tional Central University of Taiwan, the Space Telescope Salzer J. J. et al., 2001, AJ, 121, 66 Science Institute, the National Aeronautics and Space Ad- Sanduleak N., Pesch P., 1984, ApJS, 55, 517 ministration under Grant No. NNX08AR22G issued through Sanduleak N., Pesch P., 1987, ApJS, 63, 809 the Planetary Science Division of the NASA Science Mis- Sanduleak N., Pesch P., 1989, ApJS, 70, 173 sion Directorate, the National Science Foundation Grant No. Sanduleak N., Pesch P., 1990, ApJS, 72, 291 AST-1238877, the University of Maryland, Eotvos Lorand Schawinski K. et al., 2014, MNRAS University (ELTE), the Los Alamos National Laboratory, Schlafly E. F., Finkbeiner D. P., 2011, ApJ, 737, 103 and the Gordon and Betty Moore Foundation. Sobral D. et al., 2009, Monthly Notices of the Royal Astro- Funding for SDSS-III has been provided by the Al- nomical Society, 398, 75 fred P. Sloan Foundation, the Participating Institutions, Sobral D., Best P. N., Matsuda Y., Smail I., Geach J. E., the National Science Foundation, and the U.S. Depart- Cirasuolo M., 2012, MNRAS, 420, 1926 ment of Energy Office of Science. The SDSS-III web site Stroe A., Sobral D., 2015, MNRAS, 453, 242 is http://www.sdss3.org/. Thuan T. X., Izotov Y. I., 2005, The Astrophysical Journal SDSS-III is managed by the Astrophysical Research Supplement Series, 161, 240 Consortium for the Participating Institutions of the SDSS- Troja E. et al., 2017, ArXiv e-prints III Collaboration including the University of Arizona, the Tully R. B., Rizzi L., Shaya E. J., Courtois H. M., Makarov Brazilian Participation Group, Brookhaven National Lab- D. I., Jacobs B. A., 2009, AJ, 138, 323 oratory, Carnegie Mellon University, University of Florida, Ugryumov A. V. et al., 1999, A&AS, 135, 511 the French Participation Group, the German Participation White M. et al., 2011, ApJ, 728, 126 Group, Harvard University, the Instituto de Astrofisica de Wright E. L. et al., 2010, AJ, 140, 1868 Canarias, the Michigan State/Notre Dame/JINA Partici- York D. G. et al., 2000, AJ, 120, 1579 pation Group, Johns Hopkins University, Lawrence Berke- Zacharias N. et al., 2010, AJ, 139, 2184 ley National Laboratory, Max Planck Institute for Astro- physics, Max Planck Institute for Extraterrestrial Physics, New Mexico State University, New York University, Ohio State University, Pennsylvania State University, University ACKNOWLEDGEMENTS of Portsmouth, Princeton University, the Spanish Partic- We thank the anonymous referee for their rigorous and help- ipation Group, University of Tokyo, University of Utah, ful feedback. This work was supported by the GROWTH Vanderbilt University, University of Virginia, University of (Global Relay of Observatories Watching Transients Hap- Washington, and Yale University. pen) project funded by the National Science Foundation Based on observations obtained at the Hale Telescope, under PIRE Grant No 1545949. D. H. Reitze gratefully ac- Palomar Observatory as part of a continuing collaboration knowledges the support of the NSF (award PHY-0757058). between the California Institute of Technology, NASA/JPL, We thank Adam Miller for helpful comments. Yale University, and the National Astronomical Observato- The Intermediate Palomar Transient Factory project is ries of China. a scientific collaboration among the California Institute of Technology, Los Alamos National Laboratory, the Univer- sity of Wisconsin, Milwaukee, the Oskar Klein Center, the Weizmann Institute of Science, the TANGO Program of the University System of Taiwan, and the Kavli Institute for the Physics and Mathematics of the Universe. This research has made use of the NASA/IPAC Ex- tragalactic Database (NED) which is operated by the Jet Propulsion Laboratory, California Institute of Technology, under contract with the National Aeronautics and Space Administration. This publication makes use of data prod- ucts from the Wide-field Infrared Survey Explorer, which is a joint project of the University of California, Los Angeles, and the Jet Propulsion Laboratory/California Institute of Technology, funded by the National Aeronautics and Space Administration. The Pan-STARRS1 Surveys (PS1) and the PS1 pub- lic science archive have been made possible through contri- butions by the Institute for Astronomy, the University of

c 0000 RAS, MNRAS 000, 000–000 CLU I: Galaxy Catalogs from Preliminary Fields 27

Preliminary Fields Properties Field RA DEC Filter Obs-Date 5σ Detection EW Limit EW Limit center center Limit Σ=2.5 Σ=5 (name) (J2000) (J2000) (name) (AB mag) (A)˚ (A)˚ p3967 13:01:56.1 28:07:30.0 Hα1 2012-05-09 18.59 ± 0.12 7.74 16.27 Hα2 2012-05-09 18.73 ± 0.12 7.97 16.76 Hα3 2015-06-02 18.63 ± 0.17 9.41 19.78 Hα4 2015-06-02 18.54 ± 0.14 9.44 19.85 p4063 14:06:35.6 30:22:30.0 Hα1 2014-02-15 18.61 ± 0.18 7.06 14.77 Hα2 2014-02-15 18.61 ± 0.18 7.27 15.22 Hα3 2015-02-04 18.87 ± 0.19 9.27 19.47 Hα4 2015-02-04 18.92 ± 0.19 9.30 19.54 p4064 14:22:25.1 30:22:30.0 Hα1 2014-02-15 18.43 ± 0.18 8.09 17.03 Hα2 2014-02-15 18.61 ± 0.17 8.33 17.54 Hα3 2015-02-04 18.88 ± 0.18 7.87 16.42 Hα4 2015-02-04 18.96 ± 0.20 7.90 16.47 p4070 15:57:21.8 30:22:30.0 Hα1 2013-03-26 18.65 ± 0.19 6.17 12.85 Hα2 2013-03-26 18.97 ± 0.19 6.36 13.24 Hα3 2015-07-01 18.65 ± 0.13 6.75 13.99 Hα4 2015-07-01 18.62 ± 0.14 6.77 14.03 p4071 16:13:11.2 30:22:30.0 Hα1 2013-03-26 18.95 ± 0.20 6.05 12.59 Hα2 2013-03-26 19.07 ± 0.20 6.24 12.97 Hα3 2015-07-01 18.45 ± 0.13 8.27 17.29 Hα4 2015-07-01 18.56 ± 0.14 8.30 17.34 p4072 16:29: 0.7 30:22:30.0 Hα1 2013-03-26 19.00 ± 0.20 5.91 12.27 Hα2 2013-03-26 19.07 ± 0.20 6.08 12.64 Hα3 2015-09-28 19.28 ± 0.28 9.51 20.00 Hα4 2015-09-28 19.06 ± 0.27 9.54 20.07 p4073 16:44:50.1 30:22:30.0 Hα1 2013-03-26 19.01 ± 0.21 5.79 12.02 Hα2 2013-03-26 18.91 ± 0.19 5.97 12.38 Hα3 2015-09-27 18.86 ± 0.13 7.80 16.27 Hα4 2015-09-27 18.84 ± 0.15 7.83 16.32 p4074 17:00:39.6 30:22:30.0 Hα1 2013-04-29 18.88 ± 0.19 6.65 13.87 Hα2 2013-04-29 18.96 ± 0.18 6.85 14.29 Hα3 2015-08-01 19.01 ± 0.16 8.05 16.82 Hα4 2015-08-01 19.12 ± 0.16 8.08 16.87 p4075 17:16:29.0 30:22:30.0 Hα1 2013-04-29 18.75 ± 0.20 7.27 15.23 Hα2 2013-04-29 18.88 ± 0.17 7.49 15.69 Hα3 2015-08-01 19.01 ± 0.14 10.39 21.95 Hα4 2015-08-01 18.89 ± 0.24 10.42 22.02 p4095 22:32:58.0 30:22:30.0 Hα1 2013-05-24 18.33 ± 0.17 6.58 13.72 Hα2 2013-05-24 18.47 ± 0.18 6.77 14.13 Hα3 2015-09-29 18.77 ± 0.15 7.98 16.64 Hα4 2015-09-29 18.78 ± 0.17 8.00 16.70 p4096 22:48:47.5 30:22:30.0 Hα1 2013-06-21 19.00 ± 0.31 6.74 14.07 Hα2 2013-06-21 19.14 ± 0.29 6.94 14.49 Hα3 2015-10-26 18.61 ± 0.15 6.82 14.15 Hα4 2015-10-26 18.57 ± 0.16 6.84 14.20 p4098 23:20:26.4 30:22:30.0 Hα1 2013-06-21 19.05 ± 0.29 6.52 13.61

c 0000 RAS, MNRAS 000, 000–000 28 D. Cook et al.

Table Continued Hα2 2013-06-21 19.03 ± 0.30 6.72 14.02 Hα3 2015-10-29 18.67 ± 0.13 8.15 17.01 Hα4 2015-10-29 18.70 ± 0.14 8.17 17.07 p4192 00:41:22.8 34:52:30.0 Hα1 2014-02-14 18.41 ± 0.19 6.92 14.47 Hα2 2014-02-14 18.58 ± 0.17 7.13 14.91 Hα3 2015-09-26 18.84 ± 0.19 7.41 15.43 Hα4 2015-09-26 18.85 ± 0.20 7.44 15.48 p4193 00:57:55.9 34:52:30.0 Hα1 2014-02-14 18.50 ± 0.16 6.68 13.94 Hα2 2014-02-14 18.59 ± 0.17 6.88 14.36 Hα3 2015-02-03 18.33 ± 0.13 7.22 15.02 Hα4 2015-02-03 18.65 ± 0.16 7.25 15.07

Table 4: The basic properties of the 14 preliminary fields analyzed here. The first three columns list the name and central coordinates of the 14 preliminary fields followed by the basic information of the four Hα fitlers in each pointing. The information listed for the four Hα filters are from left-to-right: the filter name, the observation date, the median and standard deviation 5σ detection limits for the 11 chips in the pointing, the EW limit for the Σ = 2.5 and Σ = 5 limits set by the scatter in bright continuum sources as seen in the color-magnitude diagrams of Figures 5-8.

c 0000 RAS, MNRAS 000, 000–000